Thanks for the replies.

lnewqban, I used essentially the same method as described in the "If It Flies" column in MA. I broke the fuse into multiple sections and then did a weighted sum of all the components including the wing and stab (stab effective area was reduced for downwash). I think the disagreement with the cg model occurs because my calculations assume that the fuse sections have the same lift characteristics (for a given area) as the wing. I would think the lift slope for a fuse section would be much flatter than for a wing section. Seems to me that the a fuse section contribution needs to be adjusted downward to account for a much flatter lift slope. But what adjustment?