[1st point]
We have:
\vec {Δp} =m( \vec Δv )+ ( \vec v )Δm, Δm=/=0, Δm="change in mass".
We can hypothesise that this "transfer of mass"Δm[x]+"change in internal mass distribution"Δm[r], in general.
So, we may have Δm=0 superficially, but Δm=Δm[x]+Δm[r] , in reality. (x='linear measure'...