I think there's a lot of benefit to science as a methodology or process even in contexts where science might not seem readily applicable. The author takes as an example issues of justice. While it's true that there's something fallacious about trying to reason about oughts from scientific principles there is I think a lot of value in grounding these discussions in the scientific process. The reason for that is that scientific principles force people to avoid 'language games' or rationalisations and provide a common formalism to make 'progress'.
People often use 'scientism' as a pejorative for a sort of cold, broken calculator like mentality, but a scientific account of justice, grounded in biology, evidence about punishment and so forth would almost certainly be more humanistic, forgiving, and productive than say, religious debate or ad hoc moral justification, which has led to some of the more cruel treatments in our history.
The author also I think makes some basic mistakes about the nature of science, she says for example:
"But too often those elegant mathematical models turn out to be based on assumptions about “rational economic man” true of no real-world economic actors."
This is true, but not really a problem. As Milton Friedman put it: "Truly important and significant hypotheses will be found to have "assumptions" that are wildly inaccurate descriptive representations of reality, and, in general, the more significant the theory, the more unrealistic the assumptions (in this sense) Why? Because such hypotheses and descriptions extract only those crucial elements sufficient to yield relatively precise, valid predictions, omitting a welter of predictively irrelevant details. Of course, descriptive unrealism by itself does not ensure a "significant theory"
Believing in the 'rational man' is no mistake of scientism but like believing in a perfect sphere. Somewhat absurd and evidently not found in the real world but nonetheless useful as a tool to form theories.