No. In fact, the math that I showed you is valid for incremental sequencing; it does NOT require that a long sequence occur in a single step.
For probability calculations, in fact, the chances of getting a final long (unaided) sequence correct are the *same* whether such a sequence if formed all at once or in stages compromised of the same or other processes.
What do you mean by "correct"? Are you asserting that if evolution were rerun, it would have to arrive at the current state to be considered correct? How do you determine or define correctness?