Welcome to the Q&A Forum

Google and other search engines know about XML files.  The reason I don't necessarily say indexed is because you don't see them really popping up on a SERP page.  Robots know the difference between a straight xml file and an rss feed because of the technical markup and structure of the actual file.

For example a very common use for Google and xml files is sitemaps, which you can give them the sitemap feed in Webmaster tools so they will know about the file.  If you did a search for www.rootdomain.com + sitemap file you would never see an .xml file actually popup.  You would need to visit the url directly.

I believe that canonical tags are going to be the answer for you.  This would be a per page thing you would need to add, but if you add http://www.kitchensolvers.com/services/"/> on the Services page even if a spider navigates through one of your franchisees sections of the website and this happens: http://www.kitchensolvers.com/RochesterHills/services/ then the canonical tag will tell the search engines that they really need to see this page as http://www.kitchensolvers.com/services/

Depending on how much resources you have you may be able to find a developer who could figure this out and do it dynamically or manually is another option if you don't have a developer available.

The other thing I find useful when there is puzzling links to our clients websites is scanning through Google Webmaster Tools.  Go into the links section and it will find show you the exact url the link is located on.  This does though assume that the link on cnn.com is crawlable, indexable and has been displayed.  If the scenario is happening that EGOL mentioned then this recommendation will not help.

These returning visitors are not necessarily coming from the same source.  These users may have come through organic, direct or referral sources.  Then on their 2nd+ visit they went through an Adwords Ad.  The same user may have also clicked on multiple Adwords Ads as long as they do it naturally and the click doesn't get filtered by Google's click fraud algo.

SEOMoz's toolbar focuses on their Page Authority & Domain Authority stats, which emulate the Google Page Rank score to a degree.  I would recommend focusing more on PA & DA and not Page Rank.  I also have the SEOBook toolbar installed and that is where I see Google PR.  You could also use Google's toolbar.

Of these 3 options you laid out I would recommend adding new pages.  You can always find other content about your company/services/products that is not posted on your website and recycle it to a degree.  Have you guys ever thought about implementing a blog into the website.  As far as a great user friendly way to continue to product content a blog is great.

If your classifieds are categorized in any way then after a listing is expired I would always 301 redirect it up to the relevant category, but I think it would be more relevant to 301 it to the relevant category than trying to decipher a similar specific listing.  I would find a way to dynamically after a listing expires it implement a 301 redirect.  I don't believe canonicalization is going to help here unless you have other duplicate content issues with the site.  I would not worry about the size of an .htaccess file because it doesn't impact anything negatively.