Two issues I see with your position:
First, scaling layer-1 and implementation of layer-2 solutions is not an either/or proposition. They can be (and are) developed in parallel and one can and should complement the other. Even if ETH2 manages to increase transaction capacity by 5-6 orders (which is quite a tall order) of magnitude there is a strong likelihood that a lot of that will be taken by dapps beyond simple token transfers, and we will be back very quickly in a scenario of network congestion and (relatively) high network fees and consequently demand for payment channel networks.
Bitcoin as well, with all of its forks and different proposals has barely managed to increase its transaction capacity 10-fold with BCH. People will not wait 10-15 years to maybe get something that resembles the capacity from Lightning, so much so that the main story from most bitcoiners is that it should be seen as a store of value (not a currency).
Second,
> companies and individuals who make their money from running layer-2 services will disagree with me
Is a huuuuge ad hominem. I can tell you that there is a good number of developers (myself included, if I may say so) and companies working on this who are more driven by ideals than pure profit. The main reason that I would like to see strong layer-2 systems working (especially in ETH and with stablecoins like DAI) is that so far it is the only feasible way to have a payment network that can actually compete with credit card networks.
To put it bluntly, layer-2 is another case of "Worse is better" if we actually want to have blockchain technology as an alternative to existing financial and monetary systems. Being a purist will not help the people of Venezuela, while layer-2 projects (even custodial alternatives) could.