The logic may seem strange, but the overall limiting factor is available disk space for any video. Temporary Video (that not to be saved by rule) is mostly useless and taking up space that otherwise could be used for long-term storage of video that is of more interest (Rule Saved Video). If the program were to have two disks to put the video, then separate limits for Temporary and Rule Saved Video might make sense, but that would also impose a burden on the system to move video from one disk to the other when it is determined to save a clip. It would also impose a problem playing the video back if the system had to change which drive the video is coming from at the boundaries between Temporary and Rule Saved Video (as long as there was Temporary Video available).
As it is, all video is, from a system standpoint, just one long stream of data. It is not until disk space has run out that it is necessary to delete video that surrounds the clips to be saved. Then, all that is necessary to do is adjust the directory entries to shift the file starting point to the beginning of the clip to be saved and truncate the file at the end of the clip to be saved. This returns the disk space surrounding the clip to the pool for other use.
Look at it this way... Rule Saved Video is saved until all the allocated disk space is used up, then the oldest is deleted to make room for more video. Temporary Video is only saved a limited amount of time, thus freeing up space sooner so that Rule Saved Video can be kept longer.
The Disk Space allocation is basically how long to save Rule Saved Video and the Time Limit is how much space Temporary Video is allowed to consume. (How's that for turning the logic upside-down, inside-out and on its ear?!

)