My guess is that you are getting errors due to the power driver board. I chased all kinds of gremlins like you are having only to find out after a few service calls and spending more than necessary having my power driver board repaired that it was time to cut bait and just buy a rotten dog. My tech told me that this is the most power hungry game every built because of all the added toys etc. Once I put the new board in it solved all my issues moving forward. You also might want to check the voltage at the outlet to make sure you are getting close to 120 v ac.

http://rottendog.us/WDB089.html

The first item and the 4th item are the same thing. Likely, it's either been bashed to pieces because it was too close to the balls, or a bad cable between the sensor and driver board.
Switch #76 is the opto at the bottom of the Mini PF - likely a cut or disconnected wire.
Switch #88 is, well, the only physical switch up at the lock. Likely broken or disconnected wire.
'Rocket Kickout' .. bad coil wire? bad switch? What's wrong with it?
For the sound - board is likely needed a cap kit, or the LM1875 is fried or overheating. Either way, you need board work done there.
Could be anything. Likely disconnected wire. Could be as bad as a fried driver transistor on the Fliptronics board.

Here are a few things to help...
General - http://www.pinwiki.com/wiki/index.php?title=Twilight_Zone

Proximity switch 26 - common issue. Either the detecting coil is not close enough to the ball or one/both of the 2 pin .100 headers has failed. A common correction is to solder the two wires directly to the two boards on both ends.

I personally would stick with the OEM power/driver board. I've seen RD boards do goofy things in TZ.

Rocket kickout - could require switch adjustment. Place the game into switch test T.1. Place a ball in the rocket kicker. Got switch detection?

Sound distortion - Your LM1875 amp is overheating. Fairly common. There are a few caps that usually cause this. See: http://www.pinwiki.com/wiki/index.php?title=Williams_WPC#Popping.2C_Scratchy_Sound.2C_Very_Hot_LM1875_Heat_Sink.2C_or_Shorted_Speakers

Left flipper - examine the solder connections to the coil. Make sure you can tug on the wires lightly and they stay attached. After that, again, put the game in switch test T.1. Push the left flipper button. Make sure that TWO switch closures are enunciated.

Leveling TZ is extremely important. Make sure the inclination bubble is at the very least centered and better yet, slightly closer to the shooter.

Your tech was right. TZ is power hungry. Resets are a common irritant that can be easily remedied. But, if you don't have significant experience working on PCBs, leave it to the pros. I can help with that if you need it.

Carry on soldier. Report back.
--
Chris Hibler - CARGPB #31
http://www.ChrisHiblerPinball.com/Contact
http://www.PinWiki.com - The Place to go for Pinball Repair Info

A few things stand out.
1. Someone has repinned the female side of J101 on the power/driver board. Make sure there is a solid connection there since that is where the 5V power train originates.

2. Yours is the second TZ I've seen with the 8-Driver board located left of the MPU. Typically, they are upper right. I like the location of yours better. I'm going to move mine...someday.

3. There is a large electrolytic cap to the left of the MPU board. Looks very "hackish". Let's see what that is connected to. It could be an attempted to resolve a "reset" issue.

4. Your DMD controller board looks to have had it's HV section rebuilt. Good!

--
Chris Hibler - CARGPB #31
http://www.ChrisHiblerPinball.com/Contact
http://www.PinWiki.com - The Place to go for Pinball Repair Info

Sample games had the 8-driver board like that, instead of upper right. So, @op has a sample game, at least.

There's a hacked ROM version - or mislabeled - in there. There was never a '9.5H' released. (After someone pissed Ted off by hacking 9.4H, he stopped planning on 9.5H.) Someone changed something in that ROM and named it 9.5H. (Never even heard someone reference that version before..)

And yeah, the large cap looks to be connect up to the 5V rail. Interesting.

I saw that rom rev too and just thought, “odd”. I wonder what is different on it. I’d love to have the image.
—
Chris Hibler - CARGPB #31
http://ChrisHiblerPinball.com/contact/
http://www.PinWiki.com - The new place for pinball repair info

ACK! Please don't do this, if you can't afford to purchase a 2 pin header just PM me and I'll send you one along with its IDC for free.

That's not the end issue.
These proximity sensors use induction. Since the wires that connect the sensor to the driver are not twisted, they can pick up a lot of noise. two headers in the way adds to the noise being picked up, and shifting plugs (from vibration) and wires in said plugs will cause the driver board to pick up variances in induction enough to possibly activate the switch.

I, too, recommend not only direct-soldering the wires, but twisting them as well, to prevent cross-talk and interference.

This system is the same that's used on traffic control loops (speed sensors, occupancy sensors, traffic signals, and my experience - railroad grade crossing occupancy. Inductance works wonders - when everything is set up properly. Change something (salt on the road, for example!) and you get false readings.

I wouldn’t suggest this “hack” for any application =other= than this one. IDC connectors suck. Two pin connectors suck.

That’s suck squared.
Solder ‘em soldier. And twist ‘em too.
—
Chris Hibler - CARGPB #31
http://ChrisHiblerPinball.com/contact/
http://www.PinWiki.com - The new place for pinball repair info

Well if you are gonna do it then do it right and remove the headers first rather than soldering the wires to the header pins...this is how I envisioned the hack when I read Chris reply...thus the "ACK!"

I dont know ... there are literally millions, if not billions of them out there that have been working fine for decades. I do see that the OP has a burned Molex connector in his pics though