Unify to 𫺱 (U+2BEB1).
Suggest to add ⿱匕⿺㇉一 and 𪟽 as UCV Lv.1 due to the cognition (both the simplified form of 疑, also see the small character in the parenthesis next to the character entry). It seems that the former one is preferred by the Jing nationality (京族) in China and the latter one is preferred by people in Vietnam. I do not think we need to encode both shapes separately.

The shapes are too different to recognize as identical. If we are going to arbitrarily equate simplified components based on interchangeability, then we should apply that across the board, including 馬/马, 金/钅, etc.

The two structures have a strict correspondence, and we could treat them as only glyph variants. That is a very different situation from 馬/马, 金/钅, etc. See the colors below.

The point is not that we can derive correspondences, we can similarly derive correspondences from 馬 to 马, 鳥 to 鸟, etc. But, if we are going to say that because we can derive correspondences between glyphs that on the surface look quite different, then we should start using stronger unification that includes traditional and simplified. I don't think people want that, so the same treatment should be applied to simplified forms in languages other than Chinese used in the PRC.

The question is, how to define “quite different”? They are both the simplified forms of the same character, not the simplified-traditional relationship. What is more, the correspondence works between strokes but not components; however, 馬 obviously has a lot more strokes than 马, you cannot simply establish such a correspondence.

For example, do you think that

04658

鼠 208.9.1

UTC-03486

TS 19 · IDS ⿺𣆨𮧓

and

04657

鼠 208.9.1

UTC-03485

TS 22 · IDS ⿺鼠革

look “quite different”?