Fees of $30 is not far of the current price and no one with a brain cell is doing micro-transactions using Bitcoin just now
unless you consider $100 as being micro payments instead of a few cents so I cannot agree with you here.


Playing with block size and timings like ETH has done can improve the network and in theory Segwit should had
helped but in practice fees continued to increase since its introduction so we must all be using the currency wrong.

What the developers are doing here is saying the car engine is broken, you need a new one that runs on "Banks"
instead of "gas" but really the old engine just needs fixing and they need to apply distributed processing to the
block-chain itself and brake it down into manageable sized chunks.

Block headers are quite small and are connect like a linked list to each other but instead of containing the block
itself it could just become a pointer to a collection of nodes that process the block data or we could have block ranges
in the header like is used in Whois

This approach for sure would turbo boost the network way above IOTA speeds but would still be kicking the can down the road
because the headers would become too big in time if we all start sending $0.01 to view VIP web-pages but all
spending over say $0.99 should be on the public BC or we may as well not have one in the first place and if other
networks can present a public view of a large block-chain then so can we without using banks

Here is the white-paper https://lightning.network/lightning-network-paper.pdf
needs chucking in the bin if you ask me and I am far from being alone in saying this