So bounce, not buzz. Hi amplitude low frquency (freq determined by spindle rpm) can be dealt with using mass. The stiffness, not mass, of any connecting components to the large mass is very important. A stiff structure can translate kenetic energy to a large mass with little deflection. So, a large mass placed across the bottom of lathe legs can work, and how well depends on the stiffness of the connecting components. However, The closer the large mass can be placed to the origin of the energy, ie the spindle centerline, the less the length of the torque arm, thus reducing the forces and required stiffness. So yes, more massive spindles, head and tail stocks, and bed rails all contribute positively.
But, spindly legged lathes can be helped tremendously with properly weight placement and stiffening of the legs - I had one and was able to greatly improve the bounce issue by tying the legs together with 1/2” plywood, making the stand a torsion box (essentially boxing the stand in). A shelf 1/2 way down the legs loaded with a couple hundred pounds provided more mass which the torsion box structure translated the forces to.
As for spindle bearing loads, more spindle bearings are probably trashed by thrust loads introduced by the operator, ie beating on something inline with the spindle (most use caged ball design that is great for radial loads but have lower thrust capability, and not intended for impact thrust loads). Very high thrust generated by really cranking in the tailstock can take them out over time as well. Depends on the specific lathe and mfr.
