Based on what you have described so far, all four possibilities remain: (a) you and your brother are both right, (b) you are right and he is mistaken, (c) you are mistaken and he is right, and (d) both of you are mistaken.
It is possible that you and your brother have come to rely on different cues to the perception of imaging / instrument placement, and your past disagreements about it speak to that. And it is possible that the linearity of the Accuphase amp with respect to (frequency-dependent) speaker conductance versus the nonlinearity of the March Audio amp, together with the steep impedance vs. frequency dip of the Magicos, leads to different perceptions by your brother and you (due to differing ears and auditory systems and history and self-learning/training). A "scientifically plausible explanation" would be helpful only insofar as warding against dismissing possibility (a) as implausible or highly unlikely.
That you and your brother disagree on the status of imaging of your system, as also considering the many arguments about "imaging" and "soundstage" on ASR and other forums, shows that you are dealing with a "subtle" effect, as opposed to a "gross" sensory effect. With a gross sensory effect, there is much less need for caution in drawing conclusions, and indeed we could not function in day-to-day life without trusting our senses.
When it comes to subtle effects, the fuzziness of the limits of perception and the enormous complexity, inconstancy and variability of auditory perception are some the reasons why the war between subjectivists and objectivists will never end (the enormous suggestibilty of brains being another major reason). The previous responses to your OP have listed a large number of possible confounding factors, among them possible bias inherent in sighted testing. So far, both you and your brother have judged based on sighted testing which is vulnerable to psychological bias leading to biased perceptions. I can believe DBT is a PITA, and I am not suggesting a DBT by either of you (far easier to just agree to disagree), but if done it would help remove some confounding factors when it comes to subtle perceptions. As things stand, it seems to me that it would be methodologically questionable and premature to pick possibility (a) over the others, even if one could come up with a scientifically plausible scenario for it.