On IEM soundstage, I agree that there isn't space to create natural occurring soundstage. That said, also From personal experience (FPE? FPEx?), they can create very focused and pinpoint accurate imaging and reference flat FreqResponse, and there are a Lot less shape variables for a surround HTRF to account for with drivers in-ear, so soundstage could be processed into the sound well, and surround could work well with a broader range of individuals (if it takes in-ear driver positioning into account).
I just don't need the isolation of my ER•6i much anymore.