For reproduction of the MMB, cells undergo classical bacterial division where the genome is replicated and the cell divides. The difference is that this is happening in a sphere so the sphere doubles in size and then the entire sphere splits into two. This is commonly considered clonal reproduction (although there is an argument for if "clonal" truly exists...)

Because of this reproduction, it would take only several generations for most genomic heterogeneity to be diluted out and the whole consortium (i.e., single MMB) to become clonal. What this suggests is that MMB are purposefully maintaining there genomic heterogeneity, likely for evolutionary purposes.

As far as a comparison to the Portuguese man o' war, it is similar in many respects. The theory is that MMB have a division of labor, meaning one cells has a job of say, metabolizing acetate, and its neighbor might have the job of storing carbon, and its neighbor has the job of reducing sulfate for energy. This would mean MMB have specialized cells that have a specific function. There is evidence for this but it has yet to be proven. The cyanobacterium Anabaena has specialized cells called heterocysts that fix nitrogen, a difficult job in the presence of oxygen which is a by product of the other cells. This is a clear example of a cellular division of labor in Bacteria. But Anabaena can exist as a single cell, meaning it is not obligately multicellular, which MMB are.

I do not think there is much of a relationship to hydrazoans. MMB likely developed their multicellularity from an incomplete cell division that resulted in the daughter cell staying attached to the mother cell, eventually resulting in the current organism. Multicellularity is not monophyletic and has evolved independently across the whole tree of life. Some research cited within our article shows that a single mutation to a single gene can cause multicellular traits to arise in bacteria.