'One tool to rule them all' probably doesn't exist, and probably isn't even a good idea. For monitoring, Nagios is decent, but you might want to take a look at the Planet DevOps blog... excellent coverage of the state of open-source monitoring software. Newer options such as Graphite, Monit, etc. provide a lot of functionality, and are becoming easier to integrate.
The place you should focus, IMHO, is on supplementing your areas that seem to be weaker: documentation and management. Chef is a wonderful open-source tool for configuration management. It was originally very *nix-entric, but Windows support was added to recent releases. Definitely put an end to one-off scripts.
As for documentation, your documentation and collaboration tools should really be the same, so that documentation can evolve rather than become outdated, etc. A good CMS should allow you to do both. You could also take a look at ChiliProject (a fork of Redmine) which combines features such as an issue tracker, wiki, source repository management, document management, and more into an integrated and easy to install/manage project. I use this every day and it makes life MUCH easier. It also exposes a REST API as well as a plugin architecture that makes it easy to integrate with other systems.
It might take a few tools, but you can easily combine them to have a solid, collaborative system to document, monitor, and manage your infrastructure. But rather than trying to a single, monolithic tool, build the toolkit appropriate for your situation.