He did. Still, he's not entirely correct. Legacy profit sources in the US do vary, but tend to be primarily from 1) wholesale and retail financing (loans, leases,ancillary products such as extended warranties/service contracts) and then 2) parts. Despite his generalization his point is actually strengthened by added information. The F&I business is mostly non-profit for Tesla, and is inversely related to creditworthiness and education of customers, so Tesla is ill-positioned for high F&I income anyway.
The message in this is not complex: Tesla profits derive almost entirely from new vehicle sales and new TE sales. nearly every service category, even leases and collision repair are structured to be essentially breakeven. Even the Supercharger network is structured to be long-term breakeven.
There are other categories, though, that are different and are the Tesla magic equivalent to those calegories so crucial to other OEM profits. Those are 1) paid upgrades, 2) Autopilot (i.e. EAP & FSD), 3) Software + Connectivity. These are just beginning to be evident. Over time they'll become increasingly material, especially as subscription services in all three categories begin in earnest.
In one traditional F&I category Tesla is now experimenting. It remains to be seen how Tesla Insurance plays out but there is huge potential worldwide for auto insurance that ties directly to actual vehicle behavior patterns. That has major potential to totally alter traditional vehicle insurance actuarial processes, but using facts rather than inferences.
So, I think Elon was using shorthand for traditional auto industry vs Tesla, skipping the major Tesla advantage: direct distribution. That enables huge other benefits.
As the fleet ages the Tesla benefits will actually grow from all those factors, just not so much from the pure parts play of other OEMs.