Both of the above are terrible advice because they just reflect the biases of the posters and lack any semblance of nuance.
Bottom line if you have a first author Nature paper and a PhD, you're good to go from a research standpoint full stop, whatever you do on top of that should be directed by your interests. If you have no experience with neurosurgery clinical research, I think it's a good idea to get some as that's likely the only research you'll be able to pursue during your junior residency and you should learn how to do it for that reason, not because you need to demonstrate interest in the field. It certainly will not detract from people's perception of your PhD, because again, first author Nature paper. Even if you were someone who wasn't necessarily interested in neurosurgery when you started your PhD and then got interested during your clinical years, so what? Programs want smart people who have shown they can be productive and have come to a mature decision about their desire to pursue neurosurgery. Similarly, if you think your ultimate career goals would be best served by pushing out another high impact basic science paper and it's a choice of one over another (though it really shouldn't ever be), then do that.
Where I would push back a little is how do you know you're interested in neurosurgery if you haven't done any clinical rotations and your research is in a totally different field? While I don't think there's anything wrong with coming to neurosurgery later and if you were productive in your PhD years that will help you regardless of field, I have lost count of the number MD/PhD wannabe neurosurgeons who have approached me at meetings/med school functions etc. who haven't set foot in an OR or done a clinical rotation but are absolutely certain they want to do neurosurgery; most of them end up in another subspecialty. This isn't something you'll be interrogated about in interviews if you're PhD is in a totally different field - that happens all the time and you'll have done clinical rotations and subi's before you interview, but it's something you should think about as you structure your research trajectory for the next several years.