I decided to write a tool that would aggregate data from web sites as one surfs the web, kind of like a browser with some extra functionality. It soon became apparent that the <mx:HTML> component didn't have a simple "source" property where I could see all the HTML from the website.
Using the last one along with the innerHTML property, we can get the source of the body element.
I'm sure you could get the entire document source if you wanted to examine the page head, etc. But these weren't useful for me.