Innovation is not burning a gigawatt of mining power to process less than 4 transactions per second.  Especially when we could increase that capacity by a factor of eight with almost no additional costs.

In economics, sometimes it's helpful to try and calculate the true cost of something by factoring out the subsidies. In this case the subsidy is the block reward and the cost is paid by investors/speculators.

if users actually paid the full cost of their transaction now, it would be several dollars each. That's a horribly inefficient system and not one worth investing in, IMHO.  The fact that it could potentially be much more efficient if some minor changes were made is irrelevant if there is no process for making those changes.

The governance model needs to change, so until Bitcoin Classic or something like it achieves a clear majority of support by nodes and miners, we have to assume the rough consensus mechanism of a small minority having effective veto power is going to continue, which means nothing is going to get done. Blocks will fill up. Fees will increase. There's really only two outcomes without a higher or removed max block size: stagnation or network congestion failure.