Wow, talk about a response! First, thanks to everyone for your input - it is all appreciated.
Let me touch on a couple of things I mentioned initially:
1) DTC 301 and 304 do refer to misfires in cylinders #1 and #4 respectively. My first thought also was a bad coil. The coil for cylinders #1 and #4 was replaced with a new (known good) unit - the problem persisted. Subsequently, I swapped each and all of the coils with new units not once but twice; each time the car was test driven with the codes ultimately reappearing. The ignition control module was swapped not once but twice with two known good units, also with the original coils as well as the two new sets of coils. Same result of either DTC 301 or 304, also 401 would occasionally surface.
2) The EGR has been replaced twice with new units. DTC 401 still codes occasionally.
3) The plugs are new, properly gapped, and have been checked by switching to non-paired cylinders. If the problem were one or more bad plugs, the problem should have travelled to a different cylinder when the offending plug was moved. I even removed all of the plugs and randomly reassigned them to different cylinders. However, the problem did not move. The plugs were even replaced with another new set. Still coding DTC 301 or 304.
4) The same thing which was done with the plugs was done with the spark plug wires. The wires are newer (maybe 10k on them) and were returned to the retailer under warranty and replaced. Still coding DTC 301 or 304.
5) My OBD scan tool did not show any out of spec data during real time monitoring or during capture mode. However, I will mention my scan tool is not as advanced as GM'* (obviously).
6) The GM service tech was explaining to me that he made the symptoms go away by tapping the injector (mysterious "tap test") for cylinder #4, but he was pointing at cylinder #5 while talking. I corrected him that he was pointing at the wrong cylinder, but he insisted that he was pointing at #4. During his entire explanantion (not very enlightening) he referenced his scan tool data for me, clearly showing misfires in cylinder #4. For everyone out there who knows the position numbers of cylinders on the 3800, you will all agree with me that there is no way you can mistake the two. Sorry, but to make a mistake like that in front of a customer who knows a little bit is unforgivable. It didn't help that he was prescribing: full fuel system flush, replacement of ALL INJECTORS, and "CLEANING"
OF INTAKE MANIFOLD GASKETS. He lost all credibility with me.
7) It is also important to mention that the car has NEVER presented with a "no-start" situation nor a sudden, unexplained dying of the engine. This leads me to believe the cam and crank sensors are still good. When the codes are not lit, the car runs like a champ: great throttle response, appropriate functioning of the TCC, and good economy.
Fuel pressure tests right where it should be according to the shop manual.
9) The engine has had a thorough visual inspection of the wiring harness and vacuum connections for any external signs of breakage or damage: nothing found.
10) The car has always been fueled with name brand gasoline from only a few local stations known to have new fuel storage tanks. The last time the fuel filter ws changed, the old one was cut open and no appreciable crud was found.
I believe I have eliminated the plugs, wires, EGR, PCV, air filter, fuel filter, fuel pressure, coils, ignition control module, crank/cam sensors, and O2 sensors as possible sources of the problem. Whatever the problem is, it is intermittent and not consistently reproducible (sounds more electrical rather than mechanical). Sometimes it codes immediately after clearing the codes, sometimes it will go weeks without a code appearing.
I am beginning to suspect the PCM as a possible culprit, since I have covered most of the common sources of problems according to the shop manual diagnostic flow charts. The other confusing thing is that even though the plugs, wires, and coils check out fine on cylinders #1 and #4, the plugs BOTH show excess carbon deposits (sooting) to an equal extent. The PCM is the next step up the ladder in the chain of events controlling combustion and firing of cylinders. If everything else checks-out fine on cylinders #1 and #4, it'* starting to look like the PCM may be malfunctioning.
The other thing that leads me to believe the PCM is at fault is that the PCM on my 1998 SLE (also with about 60K on the clock and the same mods/maintenance) was found to be faulty: it coded DTC 301 consistently and predictably under full acceleration when the RPM'* exceeded 4000 (e.g., merging on the freeway). All of the above-listed items were checked on my SLE as well, but no cause was found. The dealership tech (different guy - no longer at the dealership) went through the stepwise diagnostic listed above and came to the conclusion that the PCM was the next step (he openly admitted to being stumped). After a little soul-searching, the dealership admitted to the PCM being covered under warranty for 7yrs/70,000 miles. Has anyone noticed or heard whether 1998 was a "bad year" for PCM'* with GM/Pontiac?
However, the current tech is convinced the PCM is not at fault on the SSE and the dealership will therefore not swap the PCM under warranty. I have a spare PCM, but I am concerned over the mandatory programming of the PCM for performance. Is anyone aware of whether the swapping of a PCM on a Series II 3800 requires dealership computer programming facilities or can it be done successfully at home? If I swap the PCM'*, is the car still drivable? Can I swap the PCM'* back again or will I have lost the programming?
Thanks again in advance for your thoughts.