Thanks!
It strikes me that in one way at least this recording is sort of the opposite of OMT in that it does not try to specifically record different aspects of the soundfield and then recombine them (even though that happens anyway as by-product), instead focusing on the direct PA/SBD sound as much as possible, picked up from different locations:
Pseudo-stack taping of the PA on both sides
PAS hypercards
SBD
^
All of which are primarily direct sound contributors. What makes it work I think is that the two non SBD sources contain different amounts of all the other stuff which is of secondary importance to the direct sound (perhaps except for the guitar which was loud on-stage and thus under represented in the SBD) : The spaced omnis providing some wide decorrelated room and audience, and the PAS hypers correlated room and audience.
It's hard to argue against this when it works and I certainly won't!
However, I will say that I generally find it useful to dedicate at least one pair to avoiding this direct-sound-dominant redundancy by intentionally limiting pickup of it, so that when needed those other aspects can be brought up without also raising the level of the direct-SBD sound at the same time.. as long as one is assured of having enough of that most important direct sound in the other sources. This probably calls to mind the rear-facing microphone channel(s) primarily, yet this thought experiment also sheds some light on things:
Say I'm running a 4 channel OMT setup with wide omnis + front/rear facing cardioids. I'm offered a mono SBD feed and need to sacrifice one microphone channel to record it. Which microphone channel do I give up? As long as I know the SBD is clean, good-sounding, and all representative of the desired direct sound (a gamble perhaps!) I'd be most tempted to give up the forward facing center cardioid as that channel is an attempt at providing what the SBD feed actually provides in a better way (clear, proximate, direct source), whereas the other channels while providing some of that as well are primarily focused on recording other attributes of the soundfield. I mostly want spatialization for that dry, direct SBD, room and audience sound. The omnis alone or maybe even the rear facing channels alone should be able to supply that.
I've not done so, and it would take cajones for a taper to try it, but this is actually an argument to try recording just the stereo SBD + a pair of rear-facing microphones if limited to recording 4 channels. Next time I get a SBD and have stereo rear channels going I'll try a mix of just those two parts to determine how well this works.