I just don't necessarily like when people say that "X" is the only way to go and there is no other alternative because that is just the way it is. Even though this problem doesn't seem like much right now, the more people start using bitcoin, the bigger this problem becomes... so it's better to fix it now... am I not right?
Okay "X is better than Y", where X represents Segwit and Y represents Classic/2MB blocks. That's the right way to see it then I guess. You're right, it will become a bigger problem. However, scaling via the block size will never be able to solve this problem because it is very inefficient. Here are some quick calculations that I did a while back:
Let's say that 700 million people use Bitcoin (that's ~10% of the World population, even less), and that they all make only 1 transaction per day. Transaction size is averaging half a kilobyte today (source Bitcoin Wiki).
700 million transactions * 0.5 = 350 million Kilobyte = ~350 GB/ 144 (blocks per day) = ~2.4 Gigabyte per block. This is only if they make a single transaction per day (which is unlikely, as the average would be much higher with adoption on this scale).
350x365 = 127 750 GB per year (i.e. 127 TB). This can be avoided with the second layer.