It certainly sounds as if a redirect went missing. So they're getting the naked form of the URL instead of the prettified version. You can readily test it by entering the same address yourself-- or clicking the search-engine link. See if you land on the page without getting redirected.
The two htaccess files, for old domain and new, should be almost identical. The only difference is that the final rewrite, from pretty URL to real source of content, will of course only happen on the new site. All external redirects should go in tandem.
It should be noted that if you're using any well-known CMS, google probably knows exactly what lies behind the ordinary URL. And they'll ask for it just to see if they get a page. (Disclaimer: I just made that up out of the clear blue sky.)
ALL of these will 301 to the original page, which if you remove blog/ is yet another redirect. The urls aren't being created, they actually exist because your server doesn't return a proper 404 for them. When Google finds a url it never forgets it and will return indefinitely.
Scrapers, aggregation sites and webmasters who don't double check for typos can all create malformed URLs so if your site redirects those instead of returns a 404 or 410 then you may see them appear in places like webmaster tools or analytics.
IS THIS A PROBLEM? Yes, and no. Yes because you don't want all of these 301's and malformed urls out there when they could be avoided but no because Google knows it's wordpress and sees it happening on all Wordpress sites.
There is a way to disable automatic redirects but it hasn't been shown to improve situations like yours much after the fact, if at all. Prevention is your best policy because once it happens it's hard to clean up. Just make sure your visitors have a good experience, don't fret about the phantom urls too much otherwise.