First, be careful: in the second Fetch Page module, you use <!-- START: Modules Anywhere --> as a end delimiter of the content you fetch. However, as this is an HTML comment, the parser will not find it. You need to use actual page code in any of the module's fields.
About you image retrieval, first put a regex after fetching the page, and I think something along the lines of the following regex will work fine:
^.*?img src="([^"]+).*$ replace with $1
For a safer use, you can also do it in to steps:
^.*?img src=" replace with [nothing] and ".*$ replace with [nothing again]