James, you're welcome to come along and clean up what I'm writing here. For I don't claim at all to be an expert in LXX studies. But, Enoch, here's my stab at answering your questions:
1) When I say LXX versions, I don't mean it in the same way as when we speak about the "versions" in the NT. In that context people mean the non-greek translations handed down to us (old latin, coptic, georgian, etc). When I say versions (and now I realize that I'm probably not using proper, clear wording) what I mean is that it's not correct to conclude that there is one unchanged version of the LXX. It's a translation that spans 4 (5?) centuries. Generally, it started out more functional in its translation principles. Later on it became more formal. And each translator of each book had his own patterns of translation (some more formal, some more functional, for example). With such a long time of translation and such variance within the books themselves, it's hard to not speak of versions. So when I (and I'm only speaking for myself here) say 'versions' I'm speaking about the variants within the long history of the LXX itself, not translations outside of the greek.
2) I'm taking an educated guess to your second question. The notations in the BHS apparatus when they refer to the LXX (Gothic G) are the majority of copies handed down to us, thus giving us a 'standard' text.
3) In my hard copy of the BHS it gives the LXX reading, παρθενος, but then right after that it gives us the variant, νεανις. There is a mistake in the Accordance BHS apparatus noted above. The characters are most likely correct, but instead of being in the greek font, they're in the hebrew font. So it's in hebrew letters, but makes no hebrew words. I initially thought that it was a really ugly transliteration from greek to hebrew in an old diglot. Sometimes you bump into those books where one column is the original and the next column is a transliterated wording in the receptor language (e.g. hexapla). I assumed that that's what it was in the accordance apparatus.
4) When I say "base notation" I basically mean "majority and antiquity of texts have this." But, as James noted above, evidently the LXX apparati are more precise than this.
5) I'm going to have to defer to others for the final question. When I hear "old greek" I assume it's the earliest versions of the LXX that we have. And so, I'm concluding that they could be the more (most?) authoriative determiner of the original. But, maybe the "old greek" are versions that pre-date the official 'release' of the LXX under the Ptolemies. James, do you have any more precision here? As to your last couple of questions under "5)", I can definitely say that Old Greek does not go back as far back as the authors you list. We learned greek from Xenophon, Herodous and Homer back in college. And the greek represented in this greek is substantially dissimilar to those.
Again, James, you're welcome to follow up and do some clean up work if I got some of this wrong.
And Helen, I have reported this as a "correction" from within Accordance.