In any event, you're describing one of the root causes of the problem, but not the problem itself. It's like saying "A driver isn't responsible for making the car go forward - it's only the engine that provides the propulsion". "This bit of the brain lights up when you do X" is not the same as "Humans have this problem with X, and there are ways to get around it".
Put another way: neuroscience tells us that X happens in the brain when we see Y. Now what? This on it's own has nothing to say about something like police training.
The reason that I made my comment is that the article tries to say that a deficiency in facial recognition of persons of another race is a psychological - that is, higher cognitive - effect. If it is taken as an effect of higher cognition, without a neurophysiological basis, opens up the possibility of the argument "his bias against black people made him not care enough to distinguish black people." That argument would bring open the door to arguments sociologically based biases such as being racist, when in fact there is an observed, known, studied, published, fact that the neurophysiology of the part of the brain that distinguishes faces literally cannot distinguish variations in facial features of members of a race a person hasn't been often exposed to, as a literally biological limitation that one cannot blame on sociological reasons. It has the potential to shift the conversation to a fruitless arena.
When an observation of a bias is made in the field of psychology, it is up in the air whether it is a bias in higher cognition or whether it is a bias that is due to a direct physiological effect. Psychology, as a whole, is a field that does not talk about neuroscience, or physiological reasons for behavior; it is a more top-down perspective. (Yes, there are subfields of psychology that try to engage with neuroscience, but they are not the norm.)
What you call "more technical detail" exactly addresses that distinction; the recognition of faces of people of another race which one has not been much exposed to can be directly correlated with a specific neurophysiological observation, not an observation of higher cognition which has no neurophysiological explanation.
It is a very important distinction, even if it seems subtle from the outside.
The region of the brain (at least chiefly) involved in facial recognition is the facial fusiform gyrus, and was named as such because of this observed role. However, it was later discovered that it does more than that: show, say, a car mechanic a small intricate car part that only an expert would recognize, and his facial fusiform gyrus will activate. Same thing with an expert bird watcher and species of birds.
It turns out that that area of the brain is used for distinguishing objects based on fine detail. As the article mentions, low exposure to other races at a young age will lead to "they all look the same", because there are certain sets of facial features that most often vary within members of a race.
Therefore, yes, it is true that it is not "racist" for members of another race to all look the same to you, the neural networks in this region of your brain may not have been trained to look for and recognize a certain set of features. The psychologists making the observations mentioned in the article are surely making correct observations, but the real reason isn't mentioned at all.
I would cite sources, but this knowledge is from my undergraduate degree in neuroscience from years ago, and I'm on mobile. I'll try to come back later and edit in some citations.