I'm following along, and all the concepts make sense, but I'm not coming up with the right numbers. Probably a typo. Also, the slight discrepancy between your calculated stack may have to do 1) with the small distance from the center of the top tube to the top of the top tube or 2) because that calculation doesn't take into account the fork rake. Based off of some hand-written notes and wikipedia to remind me what a hypotenuse was, I think the "true" fork length, the length that you would get if you extended a line parallel to the head tube to the level of the axle, is, get ready:
D= fork distance, the desired variable
L= fork length per spec
R= fork rake per spec
H= head tube angle

D = L * sin(H - arcsin(R*sin(180-H)/L))/sin(80-H))

I haven't checked my work yet, and I realize I haven't put my proof in yet, but my level of discomfort from not doing my REAL job is starting to surpass the enjoyment of solving this puzzle.

Hope this is on the right track...