If PeekMessage() is supposed to do the same thing that GetMessage() does but without waiting for there to be a retriavable message, why would somebody use GetMessage() after PeekMessage() has been called?

while ( PeekMessage( &msg, NULL, 0, 0, PM_NOREMOVE ) )
      {
        // Translate and dispatch to WindowProc if not WM_QUIT
        GetMessage( &msg, NULL, 0, 0 );
        TranslateMessage( &msg );
        DispatchMessage( &msg );
      }

edit: although this example would probably be fine with just the PeekMessage() without the PM_NOREMOVE.

Slight correction - add: "...without the GetMessage call."