Christian,
First of all thank you for your enthusiastic comments here and at my own essay (/1610). I am flattered to get kudos from the current top-rated essayist. (That BTW is not surprising to me, considering that your essay most resembles a journal paper proposing an advance.) Sadly I have a bit of visual trouble reading your essay, perhaps my older pdf SW did not render it right (it has the scratchy look for me of "Ghostview" altho I downloaded the file itself.) I do note first that you recognize the importance to information issues of trying to resolve controversies over BH information representation and barriers. (This is something I remember as issue from back in early days of hearing "black holes have no hair" - ie, only mass, spin, and charge were considered to be residual, and then on to the "black hole wars" as you relate.) It is of course necessary to try and integrate QM issues into this imbroglio, despite our not having a proper theory of quantum gravity! Then Hawking's discovery about EH radiation knocked a wrench into everything. Yet you realize he should not be hero-worshiped, brilliant as he is (even Einstein and von Neumann made mistakes.)
You make good progress in relating all the above to evolution of the Schrödinger equation, use of DDF etc. It seems you are zeroing in on your main point when you aptly note how to evade the cruder density matrix view and recover genuine SE of a pure state. (My own essay refers to how we might distinguish mixtures with the same DM, against the common view that is not possible.) This you cap up in a money quote "The result agrees with the assumption by 't Hooft that Schrödinger equations can be used universally for all dynamics in the universe", also being the second person I've read to directly try to refute Hawking's claim about loss of information from a BH. IOW, another pathway to quantum spring!
Readers may find my own essay interesting since I explain how we could find out more about quantum states or mixtures than heretofore considered possible. This involves for example repeated interactions (akin to weak measurements) of a single photon, as well as how to distinguish e.g. random H, V polarization mixtures from random R, L mixtures. My essay is indeed related to yours, in that distinguishing such mixtures should profoundly impact the BH information paradox.