For the scientific names, I admit my choices are largely subjective. I am aware that some “rules” exist regarding binomial nomenclature but is using an outdated synonym still mean my resulting book is valid as a form of scientific literature?
For example, the caspian tern, I intend to use the synonym Hydroprogne tschegrava instead of Hydroprogne caspia, or for the Pallas’s gull Larus ichthyaetus instead of Ichthyaetus ichthyaetus
If you think the junior name is valid over the older name for various reasons (the holotype is based on inadequate or non-diagnostic material, the older name is preoccupied by another taxon and had to be changed, that is perfectly fine.
The rules state that you should use the oldest possible name that is still valid, unless a specific exception is made by the ICZN for the sake of scientific communication (nomen conservandum) or common practice is to use a certain name within a given field.
E.g., with the latter in paleontology we have some species that might be senior synonyms of other taxa based on circumstance (Ceratops, Troodon, Trachodon, Manospondylus, Achlysictis, Dystrophaeus maybe, Palaeoscincus, etc.), but the unspoken rule is to ignore them because the people who described them did a shoddy job and named them off of bone scraps, and we can’t reasonably tell them apart from other dinosaurs.
Honestly the only real reason I intend to use these specific synonyms is because they’re not too outdated (like for example in the 1700’s when curlews and godwits were included in the genus Scolopax ) and because they sound “cooler”
Do not do this. Aesthetic preference has no bearing on the validity of a name. As stated by Wedel (2009):
Finally, I beg forgiveness from all brachiosaur lovers, that so beautiful an animal as “Brachiosaurus” brancai now has to be known by so inelegant a name as Giraffatitan.
The only possible exception to this I can think of might be Megapnosaurus, and that’s because it was deliberately coined as an insult to other scientists, which is against the ICZN rules (Appendix A, point 4).
Genus names are a bit more of a gray area, because they are subjective. About the only rule with genera is that they should represent monophyletic clades (i.e., they all represent one group more closely related to themselves than any other) and even that gets bent from time to time for the sake of reducing confusion. If you think the gulls are all similar to one another they should all be in Larus that’s one thing, but then you need to make sure that other gull lineages like Rissa are in there to ensure it’s monophyletic. And it should be done for biological or evolutionary reasons, not because the name Larus “sounds cooler”.
There are “vigilante taxonomists” out there who do exactly like what you’re suggesting. E.g., I know of at least two paleontologists who either deliberately try to use the names listed above like Ceratops and Manospondylus even after the latter was ruled a nomen oblitum (as in, officially forgotten about name), or commit rampant taxonomic revision of the dinosaur family tree whenever it suits them (e.g., treat Deinonychus as a junior synonym of Velociraptor and get the raptors in Jurassic Park/World incorrectly labelled in the public mind for 32 years). Other scientists tend to view them as not very trustworthy researchers for exactly this reason.