Typically, the structure of my projects looks like this (at a minimum):
projectname
- trunk
- src
- lib
- support
- docs
- releases
The trunk
folder contains the copy of the source that I am working on right now. Also, there's a directory 'lib', which contains all the third-party assemblies that are referenced by my project.
(I reference the assemblies at that position).
The 'releases' folder contains branches of the trunk. For instance, when v1 is released, a branch is taken of the trunk so that I have a copy of the source code and all its dependencies that is necessary to build version 1 of the application. (This is handy for bugfixes. Fix the bug in that branch, merge the fix to the trunk, rebuild that branch and you have a fixed v1 of your application).
All these things go into source control. (Yes, the referenced assemblies as well). By doing so, it is very easy if another colleague has to work on the project as well. He just gets the latest version from source-control, and he (or she) has everything in place in order to be able to compile and build).
(Note that this is also true if you use something like CruiseControl for continuous integration).