[aerohead]

Without full wheel skirts, it seems like the lower two fins/ strakes/ vanes would just be embedded within turbulence.
If the flow onto the upper structure is oblique, vorticity may be introduced, which otherwise might not be present.
Drag is a function of pressure.
The panel that the 'strake' is mounted on is capable of altering pressure, however, the 'strake' itself cannot perform that function unless it somehow enhanced the performance of the panel by 'correcting' some spanwise anomaly.
I too believe that it would be hard to measure any effect outside of a laboratory setting.
Structures on mirrors are probably there to correct an aero-acoustic 'noise' issue.
Nissan LEAF headlights ARE 'turning vanes.'
Taillight forms are there to straighten flow and provide Fachsenfeld's 'burst' edge.