This is a question that cannot have a "right" answer. As mentioned in the comments, the answer would depend on the experimental system and on the specific properties you want to reproduce.
Since most experimental systems are in what is known as the "scaling regime" you probably need long polymers to begin with (I commented on the issue elsewhere).
In addition to the length, when dealing with co-polymers the effect of the sequence can be dramatic, and hence you need to be extra-careful when choosing it. Here is a recent (open access) paper that shows the phase behaviour of short-ish (20 beads) co-polymers with different sequences. It also includes examples of polymers having the same AB composition but different arrangements. As you can see, different arrangements can give raise to sensibly different macroscopic behaviours.