Maybe sb5walker will chime in on this.
Generally I only chime in when I think I can help. Kind of like the recent poster who "removed all the egr and emisions crap. what kinds of problems am i goin to run into? should i still plug in the three solinoids on the top of te valve cover even though the vacum lins are worthless?" I might be able to help the engine, (by re-hooking up the fuel pump vsv and a/c vsv, for starters) but can I help the poster? No.
People should understand that the 22re and 3vze are E
FI engines. That is to say, a computer controls the injection of the fuel and the timing of the spark. They aren't WW II German aircraft engines, or early '70s British sports car engines, which have MECHANICAL fuel injection. In our engines, like it or not, understand them or not, they have COMPUTERIZED fuel injection. It may not be a very smart computer (it isn't) but it's what we have to work with. It actually does pretty well, provided that everything is right where it's expected and all the signals it receives are reliable.
If you pull the connector on the tps, the computer probably goes right into open loop mode, where signals from the O2 sensor are ignored and the computer uses default values for fuel and spark. 90 percent of the time, these values will be too rich or too lean (usually too rich), and the spark timing will usually be more retarded than is optimal. Perhaps with the tps gone the computer just uses vafm signals, but in that case the computer will not be adding extra fuel for acceleration, and when taking the foot off the pedal, the computer won't know to cut back on fuel. Then there's the IDLE signal, which is essential to a smooth idle and which is used for a bunch of other cases, including being able to read your codes.
The ONLY way for THESE motors to run properly, unless you convert them to 22Rs, is to make sure all the sensors that the computer needs (and the tps is surely one of them) are present and are working properly and are adjusted properly.
For the OP, to make sure the tps is working properly, you not only have to check the resistance values in the idle & full open throttle positions, as per the fsm, but you also have to check that VTA-E2 resistance changes smoothly as you slowly open the throttle, with no spikes or dropouts. It sounds like the sensor was failing this test.
For arek, there is a very tight tolerance for the right tps position: it is critical because of the IDLE SIGNAL. The correct procedure is in the fsm. If you don't have one, google 1993 toyota pickup service manual
I'm sorry for the rant, but I guess I lose patience with folks who don't use the search function and don't read all the good information available in this forum. I get a little tired retyping the same thing again and again, but I'm willing to do it in cases where I'm confident I won't be wasting my time.