Power density is a function of the battery chemistry. Your point holds if we're comparing laptop batteries (where weight matters) with LiOH- batteries to a lawnmower with, say, lead acid batteries. The lead batteries are far heavier (and cheaper) and make sense on the lawnmower.

But the Bosch machine has LiOH- batteries just like the laptop. The power density is the same. Packaging might make one heavier but, again as the author points out, the lawnmower batteries probably have more robust (and expensive) packaging.

The answer is nothing more sinister than market segmentation. People will only shell out a XX% premium for a lighter lawnmower, but they'll shell out a YY% premium for a 1lb lighter laptop. So that's what the manufacturers charge.

Isn't there value in moving to a higher level of abstraction than a single neuron though? Or simplifying the basic elements for the sake of a tractable broader model?

Simulating a single atom, for example, is reasonably complex: it would be impossible with current computational resources to simulate the electromagnetic properties of a metal if we required accurate simulations of individual atoms. Yet despite ignoring what we know about the atomic models, the higher-level models are very predictive.

Not that we have such predictive, higher-level models for the brain. That's what some researchers are searching for: I'm just suggesting that those models hopefully won't require accurate simulation of individual neurons. That seems to be the pattern in other domains.