You might want to. Sound coming from an external source would strike and stimulate the outer ear structures first and that is absolutely part of how the sound in terms of scale would be interpreted by the hearing brain. It is a system after all and the internals are only part of the system (albeit major parts of the system).
I don't think that it's been superbly well established yet that over-ears activate the pinna in a way that necessarily is desirable (ie, for stereo recordings, express your own HRTF as if you were the dummy inside Harman's reference room), let alone constant.
If so you'd see a constant transfer function between two pinnae when measuring several headphones ceteris paribus, but that isn't what you see if you digitise SoundStageSolo's experiments with both the RB5000 and RB0066 pinnae, for example.
Rtings' PRTF tests were interesting but I'm not certain that they were thorough enough to be conclusive.
Anecdotally, I prefer the earbuds I'm wearing right now over most of my unEQed over-ears for whatever I'd call "spatial qualities" (a term that doesn't make a ton of sense to me for stereo recordings anyway)
. But maybe that's just sheer luck and they just happen to work well for me.