Sorry, been trying to get this right for some time before getting back to you, but not sure that I have. I'm attaching screenshots of the ?h.sphere.reg for the original, processed and the fsaverage-subjects with the destrieux-annotation. To me, it looks slightly worse after the mris_register command. Not sure that I've managed correctly with the mrisp_paint (mrisp_paint $FREESURFER_HOME/average/?h.average.curvature.filled.buckner40.tif surf/?h.sphere.reg ?h.sulc_test) but attaching a screenshot from there as well.
Would be happy to know what I could alter to get it more accurate!
Kind regards,
Carl