9mm FMJ yaws and .45 ACP doesn't. The reason is because the center of gravity on the 9mm is located toward the base of the bullet - 9mm FMJ RN is much more "pointed" than 45 ACP RN.
That is not entirely accurate.
Both 9mm and .45ACP FMJRNs can, and do, yaw.
Even extremely oblate projectile profiles like expanded JHPs, which are far more blunt than the .45 FMJRN and have a CoG even farther forward than the .45 FMJRN, can, and do, yaw. The 9mm is arguably a little bit more prone to do so, but both the 9mm and .45 FMJRNs do indeed yaw and it is not possible to predict if, when, or even to what degree, either one will do so.
Still, the greater point remains- neither the Schwartz bullet penetration model nor the MacPherson bullet penetration model has an expression for the consideration of projectile yaw as doing so would render those models largely unusable/inaccessible to anyone lacking the requisite computational capacity to perform such complex calculations- something that I think both authors would find undesirable.
The proof of these model's accuracy is in their performance and by way of example, I've compared the Schwartz bullet penetration model's prediction against a couple of well-known examples that I am sure most folks here on TFL have seen before:
Example #1:
This .45ACP 230 gr FMJRN @ 869 fps has a total penetration depth of about 65cm or
25.6 inches:
The Schwartz bullet penetration model predicts a penetration depth of
25.7 inches.
The MacPherson bullet penetration model predicts a penetration depth of
30.3 inches.
Example #2
This "yawing" 9mm 124 gr FMJRN @ 1189 fps has a total penetration depth of about 71-72cm or
28.3 inches if you "straighten out" the upward curve at the end of the bullet's path:
The Schwartz bullet penetration model predicts a penetration depth of
28.7 inches- that's pretty impressive.
The MacPherson bullet penetration model predicts a penetration depth of
30.9 inches.