Ooh it's been a long time but let me take my shot
1 - yes stereo, and linked, but it might not matter on digital input, please see my response to #3
2 - yes it is overkill. Run a test at 16 and then again at 24, then invert and compare to see if there is anything of value if you would like to see for yourself.
3 - digital should run direct and exact bits so no need to set level. (been a long time since I ran the MTII but #1 might be covered by this)
4 - you might get a sample or two of latency so they may not arrive to your workstation at the same time if you try to do all of this simultaneously, but if you compare the bits from the rips, you should see exact correlation (maybe once you line up the start or end points if they are different total length for some reason like starting or stopping at different times)
5 - Yep, that's total overkill. Again, do a test by inverting one and adding them together, to see what's left, if you wanna check for yourself.
6 - should not be any difference, but have I got a suggestion for how to find out!? Let me guess, invert and add? Yah.
:-)