^Nice work. I think this is a reasonable compromise operating under the constraint that all decks are the same height. While I personally disagree with this interpretation, abandoning a set deck height might create so much complexity that estimates are no longer feasible.
While the MSD's tend to show consistent deck heights, they are often in conflict with the model due to the timing of productions -- MSDs have to be finalized before principal photography can begin (at least on that set), while the model can be tweaked and altered well into post production, after all the footage has been shot. Probably the MOST accurate MSD is the Enterprise D, and that is because it was created a few episodes in. For this reason, I interpret it is a graphical illustration of the relative position of systems, not necessarily in scale. Its meant to provide at a glance information, not a basis for accurate measurement or calculation. For this reason, I have no problem with subtle proportions. In other words, I think and MSD can tell you HOW MANY decks the ship has, but not how high the decks are in relation to each other.
If i had to adopt a constraint, I would go with one set of heights for the primary hull (excluding bridge), another height for the secondary hull decks (or a mixture of both), and whatever the heck fits for the necks. Looking your earlier deck mapping
, both the TOS Enterprise and the refit seem to call for a different deck height for the secondary hull than the saucer.
Slightly OT -- Memory Alpha posits that the original scale for the Excelsior was 1.5x the length of the refit enterprise, then later revised up to the 1531'. I wonder, if this is the case, what feature made them increase the length only 31 feet? According to the article, the scale drawing was originally created by Nilo Rodis-Jamero.