The conjecture fails, because only a “good separated” image $X$ can be submetrizable. It should be (functionally) Hausdorff and have a (regular) $G_\delta$-diagonal. I shall cite the paper [Gru], which deals only with regular spaces.


For instance, $\Bbb R\setminus B$ is a uncountable discrete subspace of $M$, so we can weaker its topology to an Aleksandrov compactification of an uncountable discrete space which has uncountable pseudocharacter and so it has no $G_\delta$-diagonal. Also the space $X$ is not Hausdorff, because the compact subset $\Bbb R\setminus B$ is not closed in it.
But if the image $X$ is sufficiently “good separated” then the answer is positive. Namely, if $X$ is a regular space with $G_\delta$-diagonal, it is also paracompact as a regular Lindelöf space and we can apply to it


References
[Gru] Gary Gruenhage. Generalized metric spaces in Handbook of set-theoretic topology, ed. K. Kunen, J. Vaughan, North-Holland, 1984.