Suppose you want to cut the object ball 30 degrees with a center-ball shot (no follow or draw or english). To get that cut angle at medium speed, you have to set up for a fullness of hit of about 33 degrees to compensate for the collision-induced throw (of about 3 degrees). Adding inside english and will not increase the throw much. This makes it seem that inside english is less effective, but at fuller cut angles, it becomes more important.
But the effect that Marlow and Alciatore consider goes beyond that. If two surfaces are sliding past each other, there is less friction for faster sliding. On the shot without english above, the surfaces during contact are slipping. If you add inside english, they are just slipping faster. A lot of people would guess exactly the opposite, that you should always get more throw with more side spin, but that seems not to be what the balls really do.
In the simple treatment of sliding friction, the coefficient of friction is assumed to be indepent of both the speed of slipping (as long as there is some slipping) and the force pressing the surfaces together. That's the way I learned it in high school. It seems that both speed and pressure change the coefficient of sliding friction.