The typical problem I'm seeing with Win7 on multiple machines is as follows:
My laptop for example, doesn't get used often (ie. a month, maybe two may go by without it being used). When WU is triggered on it, one CPU core is used up for literally hours. Once it has been updated, the WU check doesn't take long to run.
On another laptop I have in at the moment, I saw a similar issue to what the OP described - the laptop ran with one core saturated for several hours, the WU CP UI was saying "checking for updates" the whole time, though one interesting thing I noticed (which I think I've seen occasionally before) is that Windows had finished processing updates some time ago (despite a core still being saturated), partly notable because the WU icon had appeared next to the shut down button, and partly because if the WU CP UI is closed and re-opened, it shows updates waiting to be installed and a recent 'checked for updates' time.
I don't mind the idea of the WU CP UI not being updated as it should; as UI bugs go it's easy to work around, but the CPU core saturation issue will a) chew laptop batteries up and b) stress components more due to increased load / heat produced.
No, I think what you are seeing are the inevitable consequences of an update system that lets you remove almost every update independently without killing any subsequent ones. This becomes an extremely complex system over time..
IMO while this is logical, I don't think it's correct (at least not in recent cases). For example, around the time Win7 started having WU issues (first it was the memory chewing bug, 1.7GB at worst in my experience), Vista started having them too. Now Vista is about three years older than Win7, so it would logically follow that if it was down to the sheer number of updates and inherent complexities in processing them, then Vista should have started experiencing that problem some years before Win7 did.
IMO MS accidentally introduced the memory chewing bug on operating systems with more than x number of historical updates (I've observed the bug on Win8 RTM as well, but not Win81), then they fixed the memory chewing bug only to accidentally introduce a core saturation bug. Eventually, once people have moaned at them about it enough, they'll get around to fixing this one as well, like they did the WU bug in XP that caused CPU core saturation.
After writing this I considered the possibility that the date of the last service pack for the respective OS might have a bearing. Vista SP2 was released in 2009, Win7 SP1 was released in 2011, so there still ought to be an approximate 2 year gap.