The general gist on Lepore is that her work is generally very good, but it's best on the areas that she really knows well. When it comes to branching out into other domains of expertise, the relative thinness of her account is really clear to an expert. This is not much of a criticism, because this would be the case with pretty much any historian — which is why survey books are usually written by teams, not individuals.

For example, her work on the early American republic in that book is great. That's her jam. The section on the atomic bombings is not great — it's not terrible but it is clear that her sense of it is very much second-hand, and there are things that I (someone who works on that primarily) would say differently. Sometimes this is a matter of emphasis and inclusion, sometimes this is about the specific arguments made (there is a difference between "how a historian who works on this right now" imagines it, versus "someone relying on sources that are often 10 years old" would). It reflects a sensibility on her part about how to incorporate those events into the overall narrative she is trying to pull together, but in doing so misses a lot of key points that someone who works on that topic specifically would want to emphasize. (I have no doubt that if I tried to write about something she worked on first-hand she'd see the thinness of my own account as well!)

Again, this is not meant as a harsh criticism. It's a good survey book. Survey books are hard to write. I cannot imagine trying to write such a book as an individual. But some parts reflect a much stronger mastery of the events than others. You can say this about all of her (voluminous) output — when she's working on something she really knows, you can tell. When she is branching into something she is mostly picking up second hand, an expert can tell.