Broadly, there are two kinds of published paper: review papers, which summarise and synthesise existing knowledge; and research papers, which add something new. (For an example of the former, see my paper Sauropod dinosaur research: a historical review at http://miketaylor.org.uk/dino/pubs/tayl istory.pdf and for an example of the latter, the zippily named A new sauropod dinosaur from the Lower Cretaceous Cedar Mountain Formation, Utah, USA at http://app.pan.pl/archive/published/app 100073.pdf )
You can certainly write a research paper based on your own blog-posts: the fact that original research is original research isn't negated by its having appeared on blogs before. As an example, yet another of my own papers, Neural spine bifurcation in sauropod dinosaurs of the Morrison Formation: ontogenetic and phylogenetic implications at http://www.palarch.nl/wp-content/Wedel- -10-11.pdf was adapted directly from a series of eight blog-posts linked from https://svpow.com/papers-by-sv-powskete furcation/
Do you need to handle a fossil yourself to write a paper about it? It depends. If it's a purely descriptive paper, as in our description of the awesome sauropod Brontomerus linked above, absolutely yes. Even 3d models are no substitute for the real fossil. But if you're advancing a palaeobiological hypothesis, as in the Smilodon blog-posts, then no. Your research is not about the fossils per se, but about the animals and how they lived. For that, you can rely on previously published descriptive work and photos.