A potential counterpoint to the idea of nine saucer decks is the Miranda Class. Her dorsal amidships section features the wedges to port and starboard that ought to be two decks tall if they're following the plan here:
Unfortunately, the wedges have three window rows. Although technically this only means that her aft block features decks like that, it is nevertheless emotionally unsatisfying for the decks to differ. Alas, the Miranda gets worse the farther back you go. To the sides of the aft block section are window rows that are even worse. Two additional window rows appear on the sides forward of the nacelle pylons, the lower of which seems to be some weird mid-deck setup, like a second row for the upper deck of the saucer rim.
There's a small reprieve insofar as the aft block being a bit higher at the start than the top of the third Enterprise deck, with little wraparound spindles hanging on to the lower part of Deck 2, but that only helps a little.
Aft of the pylons things get really wild, with window rows suggesting as many as eight decks for the block section (which should only be sporting four per the deck heights from the CONDITION: RED graphic).
What obviously happened is the modelers just spaced these out in keeping with the saucer edge windows, but without leaving any room between, creating a nutty situation.
Pity . . . I'll never un-see the Miranda problems again. And no, upscaling by the 20-50 percent that folks typically leap to doesn't solve the issues. You'd literally need four full decks at the saucer edge, and then to ignore every scaling against every ship including the TWoK Enterprise flyover shot. The only immediately obvious fix I see is, like the funky double-window-row thing forward of the nacelles, to conclude two rows for some decks, or inconsistent deck heights that change rapidly fore-to-aft in the block section. Their spacing aft of the pylons makes that concept baaaaarrrely squeak by, but that's one of those last-resort nuclear options.
