Me too. I have particular difficulty in appreciating how a weak wrist movement can be a major contributor to RHS when so many stronger muscles are being recruited in the forehand. IMO, analysis of static images is not sufficient to build a model of the FH. One also has to analyze the motion preceding the still image, and the motion following it.
That said, I believe the wrist does play a critical role, with flexion being a significant contributor to RHS (20% is significant). My intuitive model is that when the forward portion of the swing starts with the racquet butt pointing towards the ball, the wrist contributes power as an initial impulse, and then goes into release mode for the rest of the stroke where the contribution is passive/motion dependent. At this stage, intentional movements of the wrist/forearm are probably more control oriented, IMO. I believe the same type of analysis applies to serves, although the particulars are different.
Even with extreme grips, it seems to my eyes that flexion is the major wrist movement. However, with extreme grips, flexion will contribute proportionally more towards spin production (upward movement) than pace (forward movement), so the extreme western guys have to expend more energy to produce pace, and hitting flat is more difficult than with eastern or semi-western grips.