Like you I'd be sceptical of most repair places having any kind of test harness for such a thing these days, and therefore unable to say they'd
"fixed" it versus they'd
"opened up the box, had a poke around and done some basic electronics checks". You're needing to confirm both the hardware and the functionality, not just kicking the tyres.
All my random googling efforts last night amounted to not a lot more than what you've already got. The most extensive thing I found was your own thread on a Manta board - I won't bother post a link for you

Couldn't even see a pinout or circuit diagram. Is there a Haynes or GM service book for the Carlton/Senator/Monza? I'd searched on the "Bosch 0280220028" part number that gets banded about, before I noticed that was 2.5 and 3.0 specific and flipped over to "Bosch 0280220019" for the 2.2. I tried looking under "leerlaufregler" to see if I could pick up any of the Opel-derived mentions across Europe. Nowt though. Even searches on the usually trusty
motor-talk.de proved fruitless. Maybe at a push you could track something down to at least give you a steer using all the different part codes for item 39 here -
http://www.senator-monza.de/?MTPElERNZiMyETPElUTT1kJzUjNx0DRJVUTTZSM5ITM8NDOyEDf0EzM9EkUBBVT Aha, at least they have some wiring diagrams available it seems -
http://www.senator-monza.de/?wM9QUSE1kJzMTPElUTT1kJ5UzM9QUSF10UmMjM9EkUBBVTAnd then slight improvement as I'm typing - is this a wiring diagram for the Jetronic on a Z22E with a leerlaufregler at K40 ?

-
http://www.senator-monza.de/daten/2/2/bilder/Bilder%20Einspritzanlagen/Schaltplaene2020/Z22E%20Jetronic1024.jpg It's a big assumption on my part that's even the engine/year/ECU combination you're talking about, but at a glance the pinout seems to match the numbers on the ECU casing that I've seen in pics.
Given all of that and the fact that you're progressing down the 'verify/replace the components' path, plus the fact that you've acquired a second working ECU (until it too fails), as a longer-term pastime for my own amusement I'd adopt a different tack. If you're able to measure the known inputs and outputs on a working ECU, one might perhaps be able to reverse engineer a map if it's a linear change. Then 'quickly' knock something up with an Arduino derivative to mimic that. Far, far easier if you had
any kind of better source material to work from though, I grant you. But given the seeming cost of unknown provenance ECUs, I'd spend minimal cost and chuck some evenings at it, and give it a go
Cheers, Stu