The stereo imaging is very good, as-is, and it responds very well to tonal adjustments using M-S processing.
I have not gotten around to testing for a subtle "double-boundary" effect with these SASS-like arrays to incorporate the mild boundary effect that Paul Jacobson and I detected using high density foam and the corner placed capsules. The roles of setback distance and other baffle-related variables could very well lead to further refinements.
Here is a more detail plan for the array:
Click on image to enlarge it. Right-click to download