I recently purchased a 1998 Honda Accord EX with the 2.3 F23A1, manual transmission. It has a new timing belt, water pump and starter. It was running and driving great without any idle issues whatsoever, but started throwing a code for the EGR system. I replaced the EGR valve and it threw the same code, but would also have some hesitation when taking off at a stop light once warm. I disconnected the battery and I pulled the plenum and EGR valve and cleaned out the EGR port. Scrubbed it out and used a shop vac to pull out anything in the port. At the same time, I cleaned the throttle body, while still attached to the plenum. Replaced the upper plenum gasket, put it all back together. Plugged everything back in and verified that the TPS plug and MAP plug are plugged in to the correct sensors. Verified that all vacuum and coolant lines are hooked up correctly, topped off and bled the coolant system using the bleed valve and running until fans turn on/thermostat opens. Performed the idle re-learn multiple times, but the car now has an idle surge from about 1100-1800. I have checked everything multiple times over and I am stumped. This car does NOT have the fast idle thermos valve. With the car running, if I remove the air intake tube, I can cover the 1 and only port in the TB for the IACV bypass and get the car to idle down normal if I leave some air flow. This tells me the IACV is not being told when it needs to close. With the car running, I can hold my hand over the entire TB opening and kill the car. "I think" this verifies that there is not a vacuum leak. With the car running, if I unplug the IACV, the CEL light obviously comes on, but the car drops in RPM and has a fast bounce around 1400, until I plug it back in. Unplug MAP sensor and the car idles steady at about 1100, but this doesn't seem to tell me much.
On a fairly generic scanner, the system is bouncing between "closed loop" and "open loop".
I saw in just a couple other threads that there is a coating on the butterfly valve in the TB and if that is cleaned off, it can cause the surge. Is this an accurate fact? I can see some light around the butterfly valve, it's just crazy to think that something like this could actually cause this problem, when there is a LOT of air pulling through the bypass port for the IACV.

Here is what all I have done:
1) Replaced EGR valve (initially)
2) Cleaned EGR port
3) Bled coolant system multiple times over
4) Performed idle re-learn multiple times over
5) Replaced IACV with a cheap aftermarket (thought this may be the problem)
6) Replaced IACV again with an OEM one from a pick a part (again, thought this could be the problem)
7) Removed, inspected, re-cleaned the throttle body and re-installed with a new gasket
8) Tested IACV to verify that it was at least operational and not stuck
9) Verified that the IACV is getting 12V at the plug
10) Replaced IACV with another brand new one from autozone ($200)
11) Replaced ECT sensor
12) Replaced radiator cap
13) Replaced power steering pressure switch
14) AGAIN, verified that all vacuum and coolant lines are properly installed
15) Sprayed around all lines, IACV seal, plenum seal and intake manifold gasket from top and bottom looking for vacuum leaks and found nothing
16) Removed and re-inspected EGR, plenum, throttle body and IACV
17) TPS holds steady at about 9%

Any help would be very appreciated, as it is costing me way too much $$$ to drive my truck to work daily.