A better Coastal Command?

Two points need to be borne in mind.

Firstly, the RAF needed aircraft - almost any aircraft - to train it's greatly expanded manpower. That meant ordering aircraft that were available rather than waiting for the "best". This explains the huge numbers of Ansons, Battles, Harts and Blenheims that were built in the late 30's.

It also explains why aircraft like the Botha were ordered "off the drawing board". The Beaufort was a "stretch" of the Blenheim and was not a great improvement - but it was quicker to develop.

Secondly, Bristol took longer than was planned to get the Hercules into full scale production. Thus the Beaufort was cursed with the underdeveloped Taurus while the Botha was seriously underpowered with the Perseus. Blackburn tried to get the Hercules for the Botha and Bristol made sure that the Beaufighter did get the Hercules. It ended up as a very successful CC strike aircraft.

I give you the Banff Wing Sorties flown by RAF Banff Strike Wing 1943 - 1945, Scotland - UK

But, bear in mind that only the British and the Japanese commissioned purpose built torpedo bombers (AFAIK) - even the SM79 was an adaptation.

So you can make a good case that Coastal Command, even if it came third behind Fighter Command and Bomber Command, was still getting a lot of strategic thought and money.

Pity that the Lerwick and Botha were rubbish and even if the Blackburn B-20 looked great, the money would have been better spent on more Sunderlands!

Coastal Command was always going to need a sort of high/lo mix in modern terminology. Good as the Sunderland was it was too large and too expensive to equip a large proportion of the needed squadrons with. Unfortunately the Anson was too small and a bit too cheap. Having over 50% of your operational squadrons (and an even higher percentage of operational aircraft) still equipped with what were essentially "scarecrow" aircraft 4 years after the Anson first flew is where a big part of the problem was.
There is a large spectrum between the "Best" and a glorified "scarecrow" aircraft. Something in middle should have been do-able. It was do-able as shown by the purchase and use of several thousand Hudsons. There shouldn't have been a "need" to wait for Hercules engines. Mercury's, Pegasus or Perseus engines (or purchased Wrights/P&W engines) would have good enough for a low-end maritime patrol plane. Just a few hundred planes while waiting for the "Better/Best".
Leave the Ansons to the training commands.

Decide that the 100lb bomb was useless and that 250lbs bombs are a minimum.
Get some built in time to test and check bombsight.
Build simplified bombsight for low altitude use, should be simple as you are only dealing with 2 or 3 different kinds of bombs, not a dozen or so.
Train crews in low altitude attacks (or not so low, dropping from too low causes bombs to skip)
Fit the intervalometer so you can drop a stick or liner pattern to improve odds. 3 bomb minium (not two ) with 4 bombs preferred.

Simplest solution is to allocate several hundred Blenheims to CC during 1938/early 1939 to replace Ansons.
As problems crop up with the Botha (and to a limited extent the Beaufort) cancel Botha on the drawing board and have BLackburn build several hundred more Blenheims. Fool around with a sightly improved versions (NO EXTRA CREW).
Order a few dozen Blenheims with external torpedo carrier. At least they would have found out that the existing 18in torpedo didn't take well to "high speed" drops and start working on a torpedo that would work when dropped from the Beaufort sooner.
Think about that one for moment. Bothas and Beauforts ordered by the hundreds (if not over a thousand) in 1938 and yet it would be 1941 before they had a torpedo that work when dropped at even 150kts.
Too much procurement was based on paper plans.