I did some tinkering today and answered my own question.
The modeling method used in VSP/Vorview is ok for control surfaces if
your deflections are less than 20 degrees or so.
At angles greater than 20 degrees (flaps, for instance) you want model
the deflecting surface as a separate panel. VorLax will capture
nonlinear effects this way, but miss it with the mean camber line
method.
I'd like to share pictures of the models and the CM plot... Is there a
way that the google group can host the files?
Daniel