The CentOS project knows there are problems, but we cannot tackle those problems while a new release is imminent. The people producing updates and a new release are some of the people that take part in this discussion.
That is why the answers you are looking for during a period of high workload are probably not the answers you were hoping for. Releasing CentOS 5.3 was more important at the time.
That said, we have to learn from this. Adding more resources to te build process itself is unlikely to help a lot since there are risks involved in increasing the pool of people that can build and sign official CentOS packages.
But better planning and project management could help avoiding delays. Making it easier for people to create their own rebuilds could help with contributions. And more transparency about the process could help with getting more people involved in improving the processes and tools. In itself it would be good if there was more competition in the RHEL rebuild space again.
There is a great deal that we have to improve in the process, and not everything may be feasible given this is based on volunteering. But we should be cautious to not use "volunteering" as an excuse, nor should we accept criticism without some commitment to help.
Let's hope we can draw conclusions and find acceptable solutions before RHEL 4.8 is out :-)