We can scale the main chain now (Why is this even an issue?) and play with side chains later. Or should we just wait for n number of years and hope that Lightning Network or some sidechain solutions will appear? Seems like some Core devs think we should. :/
It really isn't so simple (which is why it is an issue) and mostly this is because the majority of mining hashrate is in China which has much poorer bandwidth internet (especially when it comes to communicating with the outside world).
Already at least a few of the Chinese mining pools use SPV mining which was what led to the recent fork after the new block versions had achieved a super-majority.
If the block sizes are massively increased then SPV mining will probably become what is being used by the clear majority which would mean that Bitcoin will be less secure and more prone to future accidental forks.
There is really nothing you can do to stop SPV mining and as the Chinese have invested millions into their operations they are not going to just stop doing that because some core devs or other people think that they should do that in order for people to buy a coffee at Starbucks (which seems to me to be an odd idea when people aren't even being paid in BTC in the first place).
Personally I don't see the urgency of any of this - Bitcoin has failed to even make inroads into the remittance market (worth billions per year) and that is by far its easiest target (a so-called low hanging fruit). So instead of trying to get it to become a replacement for the entire financial system overnight why not first wait for it to actually succeed in doing what it should be able to do better than anything else?
I am not advocating that Bitcoin should only be used as a settlement layer but I think it would make more sense for that to be the focus for the next few years.