Except that it doesn't have to. As the results of this firefight show, it's not necessary to get through the attacker's arms/hands to have a tremendous beneficial effect on the outcome. Weapons are made to be manipulated by hands, damaging an attacker's hands and/or arms are an excellent way to render him ineffective even if COM is not reached.
One would hope that an upper extremity hit would incapacitate an attacker, and in most instances it just might, but then again it also might not.
And when milliseconds count and lead is flying both ways, I would prefer a round that can still reach COM after having to penetrate upper extremity structures.
Remember Mirales ended the fight by using decisive fire AFTER he was seriously wounded in the arm. Sure it was difficult for him to use his weapon after it was mangled by a .223 round, but he did it, and he killed both bankrobbers using a handgun, one handed. Also his arm was hit by a .223 round that fragments violently within its optimal ballistic envelope, not by a decidedly "more benign" handgun bullet.
Moreover, one of the other agents that was also hit in the hand found it difficult to close the action of his revolver due to bone fragments or other detritus after he reloaded (I seem to remember), and when he was in the process of reaching for his shotgun in the back seat to re-enter the fight, he was then shot and incapacitated by a .223 round from Platt.
Lastly and most germane perhaps is that Platt himself was shot through the arm with that notorious 9mm Silvertip, before his thorax was penetrated just short of his heart, and it was only after that happened that he still went on to kill the multiple FBI agents.
Point being is that I wouldn't count on upper extremity hits to incapacitate the attacker quickly or decisively enough, and that optimally a bullet should be able to penetrate the vitals in the thorax even after having to contend with the upper extremities that can, and do, get in the way.