you completely ignore (while referring to it) the problem with the curved portion of the scan.
the lines of the boxes curve with the paper- but the letters do not
FRAUD
The area of greatest curvature (presumably due to being bound in a book) is by the left margin. There isn't much writing there, but if you look at line two for sex "MALE" you do see a little curvature. I see some curving as well in the name of the hospital "Kapiolani."
Most of the typing is either in areas where you'd expect (and find) little curvature, but on the left side there is a tiny (expected) amount of curvature that's hard to measure with the naked eye.
If there's a problem here it's more with the printed letters of form itself, which doesn't show much obvious curvature, but being smaller, it's even harder to see if they curve or not. I really doubt the forger would create the printed letters of the form with letters from different sources. The form would have been his or her given template. But if you're into microscopic analysis and have the tools you might check it out.
Look at anything closely enough and you'll notice irregularities and anomalies. Corsi takes the kind of irregularities that you'd expect from an authentic manually-typed document for signs of forgery, but a more exact, more precise document would be more likely to be a forgery.
Many of the letters Corsi complains about are in about the same place on the document: in words like Barack, Kapiolani, Male, Stanley located far on the left. The translation of a three dimensional object (a bound volume) to a two dimensional graphic may account for this. If the irregularities can be accounted for in this fashion, they aren't really irregularities.
Notice that the dimensions (width) of the letters are only really called into question where curvature is an issue. For Corsi's other points consider what happens when key strokes of different force take slightly different amounts of ink from the ribbon and are absorbed in slightly different ways by the paper.
The print world of the typographer may have been different, but everyone who's used a manual typewriter should have noticed at some point that the letters don't always come out identical on the paper even if they are produced by the same key.