Up until recently this was state of the art (or, spraying DNA really hard onto cells, or putting cells in a DNA-laced salty spike trap). All pretty barbaric methods really, and still widely used.
The fewer cells comprising the organism/thing the more likely it is to have any likelihood of a meaningful result. Which is not much at best because like the most fundamental capability of a living cell/organism is keeping out foreign DNA and not integrating and reproducing it if some does get in. So the approach is to do it on a massive scale and design in some easy way of recognizing/selecting for the successful ones i.e. if it happens in nature it's vanishingly unlikely that the zapped genotype lasts for more then a generation, just like with any other random mutation.
The likelihood of a zapped-in chunk being a major evolutionary advantage could be higher than that caused by an environmental factor, just because it can transfer big chunks of something that was once useful to someone else, versus the more point ish mutations caused by radiation etc.
In relation to the eels/lightning, the mechanism is totally plausible but the question should be more how meaningful is it in comparison to all the other ways DNA is transferred like viruses/bacteria/fungus, radiation, bad luck, etc. Generally other unicellular critters are somewhat better at sneaking in weird DNA than environmental factors because it's their trained profession or something