As Tokyo Nagoya stated in his answer, in modern Japanese, ものの is considered one word, a 接続助詞 (conjunctive particle).
However, it does in fact consist of the noun 「もの」 and the genitive case-marking particle 「の」. There is no elision occurring. Generally speaking, your thoughts on how this particle works are correct: it takes two nouns (or noun-forms) and makes on larger noun-form.
However, the genitive case in Japanese can also mark the subject of an adnominal clause, for example:
These may look like they are connecting (神)の(起こった日) and (事故)の(起こるところ), but what is actually happening is (神の起こった)日 and (事故の起こった)ところ -- the 「の」 is marking the subject of those adnominal clauses. This form of subject marking has existed since before Old Japanese.
「ものの」 is attested from the Heian Period onward, and though I was unable to find any explanation of this construction in particular, I would hypothesize that it is related to the noun-marking function.