Introducing a concept such as Air2 is placing something new and unfamiliar into the equation. Even if you show your buddy the use thereof before you descend, in the split second that something might go wrong, the human mind would revert back to what its instinctively (through drill training) been trained. In other words do expect your buddy to be looking and grabbing for a yellow second stage on your BCD (most possibly on the right side of your BCD).
That's one of the solid arguments behind teaching primary donation. A primary is a primary and it is configuration independent.
I have donated gas three times to recreational divers in OOA situations and in 2 of the 3 the OOA diver (in both cases not my buddy) immediately went for my primary with none of the air share signals or other rubbish that goes out the window when someone really needs gas right now. In the third case, the diver was my buddy, was already heavily leaning toward technical diving, was quite calm and I saw him coming well in advance of his arrival so I had my primary in hand ready to hand off to him.
If you want to do realistic training, trail your buddy by about 20'-30' at a normal swimming pace. Leave your reg in your mouth, but simulate an OOA situation and try to swim down your buddy to get gas. Do it first with full lungs, then simulate the failure after you have completely exhaled. It will be an eye opener. When you arrive at your buddy, (assuming you do so before inhaling off your reg) let me now whether you bothered getting his attention, signalling an OOA, and waited for a response. I already know the answer to that one.
The point being that introducing an Air2 into the OOA situation is going to be a moot issue. The odds are good the OOA buddy will go for your primary, and if you are really on top of the situation and see your buddy coming, you will hand your buddy your primary anyway (since primary donation goes hand in hand with using an Air 2). In either case, the buddy will NOT be looking for your octo.
Again, if you want to train realistically, unclip your octo and then have your buddy come borrow some gas sometime during the dive. Odds are the octo will end up back by your tank tank or in some other dififcult to locate place - where it will tend to end up if it comes unclipped during the dive. There is a chance that you will end up snatching your primary back (warn your buddy in advance) or you will end up going for his octo, especially if you combine that with your buddy taking your primary just after you exhaled. At a minimum, it will make you a believer in an easy to find octo - either a bungeed octo or an Air 2.
...I had to have a longer hose attached to my AIRII...
...I have had problems with free flows and slow leaks from the AIR II. This got annoying enough that I wound up doing one dive on the AIRII so I was not "wasting" the air or having to listen to it bubbling. By the end of the dive (90 minutes) my jaw was awful sore because of the bulk of the AIRII and the cumbersome nature of it being attached to the inflator hose. My neck was sore from having to cant my head to the side to breath off the AIR II.
I agree with you that an Air 2 works best with a slightly longer than normal corrogated inflator hose.
I am also not surprised your neck got sore after using an Air 2 for an entire dive. The Air 2 was never intended for extended use.
The Air 2 is also not any more freeflow prone than any other octo design, if it is freeflowing it is in need of adjustment. What is less obvious is that an Octo with the same slight freeflow is often missed due to where the Octo is located during the dive. That is again a strong argument for both an Air 2 or a bungeed Octo.
-----
It is probably pretty obvious that I prefer the bungeed octo given my technical diving, and that I think an Air 2 is fine (easy to find and consistent with primary donation and a long hose configuration) and that the only thing I really oppose is the traditional octo in the traditional "golden triangle" location and in particular combined with a "donate the octo" philosophy - as it simply does not work all that well in the real world as there are too many things that can go wrong.