True, but there is more to it then that, as even the B cells that leave the bone marrow are not hyper specific and capable of targeting multiple different antig

True, but there is more to it then that, as even the B cells that leave the bone marrow are not hyper specific and capable of targeting multiple different antigens (they are still "naïve B cells"). The second level of differentiation that can result in highly specific antibodies does not occur until B cell activation. I suspect this means the helper T cells also play a role in preventing generation of auto-antibodies (at least for T cell-dependent (TD) antigens). And once the B-Cell is activated and forms a germal center to specialize, I'm not sure what if any negative selection system is used to prevent accidentally targeting autoantigens in addition the foreign antigen. (If an autoantigen is similar to the foreign antigen, then this seems like it is rather possible.) The whole process is crazy complicated. Like most of biology, it seems like a miracle that it even works at all, much less that is seems to generally work rather well in general, despite the problems that can occur.