Basic math.

SW is an effective one time increase maybe by 100% that is it.

As for LN I would start by reading the LN paper in particular the conclusion. https://lightning.network/lightning-network-paper.pdf . Under a highly optimistic scenario they are talking about 133 MB blocks instead of 24 GB blocks, a factor of ~185x. A more realistic factor could easily be a factor of 10 or more less. So again a one time increase that is it.

The reality is that while SW and LN are likely very useful technologies in their own right they do not address the fundamental problem facing Bitcoin: How does one scale the number of transactions on the blockchain while at the same time maintaining the security of the POW and create a scarcity on the blocksize to avoid spam attacks? As for Bitcoin unlimited it fails on both the security and anti spam angles above.

If someone has a solution other than what is in Monero, namely the combination of an adaptive blocksize limit with a block reward based penalty and a tail emission, I would love to see it. In over 4 years looking at this issue I have not seen another workable solution.

Edit: Bitcoin unlimited is based upon the paper of Peter R. Rizun https://www.bitcoinunlimited.info/resources/feemarket.pdf. Again if one takes a look at the conclusion one finds that the effective penalty for increasing the blocksize is proportional to the block reward. Now what happens when the block reward goes to zero? There is no effective penalty and the blocksize increases with no constraint. So how is the POW supposed to be secured in this scenario? The other problem with this approach is that it relies on network failure in the form of a very significant probability of orphan blocks to constrain the blocksize.