Some day I hope there will be such a thing. For now, though there isn't--so you'd need to record on separate tracks, and delay the track(s) from the closer microphone(s) when you mix down.
You've definitely picked up on a major issue, though. When I first discovered what people here mean by "matrixing" (not to mention that I wish they'd chosen another word for it, since that word already has an important meaning in audio) I was a more than a little taken aback.
Sound travels at only about 1100 feet per second, so midrange audio frequencies have wavelengths in air of only a few feet, down to a foot or so. When you mix together two different versions of the same signal (or sound source) picked up from positions that are many wavelengths apart, you get a bizarre conglomeration of reinforcements and cancellations in the signals at various frequencies, depending on the position of the performers and the microphones. Your idea would go a long way toward solving that problem.
--best regards