When I tweak a system typically I do the following:
1) lessen the steps - I map out the flow chart and try to clarify the goals of each step and see if I can achieve it with less steps. If the process was written in paragraph form I break it down to bullet form steps to make it more discrete per step. If the steps are SPREAD OUT through the material I consolidate it ALL and give page references to all relevant mechanics.
Example is the Variant Rules which reduced the number of Modifiers through Logical Groupings. Applying Bounded Accuracy and Process Based (Inputs, Controls, Outputs) approach to Rolls.
2) reduce the number of options but make it more flexible - less text and option to study - use broader language or use technical categories OR increase the number of Options but consolidate them in a table or Break the Option down in to Logical Groups so that its easier to remember.
Example. In my Variant Rules of CE I had to increase the Options because of Superiority in Firepower, Strategic Advantage or Position, Numbers required Options that Divided resources and the opposing force's resources until the smaller side has the advantage of Mobility or Decisiveness.
3) illustrate or create diagrams and examples of how it works at both the player level (what they need to start the process, the inputs and results) and gm level (who will improvise or plan this).
Typically I put this in Designers Notes - SO you have a bunch of mechanics - what was the design intent, how simple or elaborate was it supposed to be. How can we scale it down to essentials how do we elaborate on it for variety.
4) benchmark against expectations.
Like I discovered the dimensions of the missile in Ce and MGT2E is similar to the patriot missile at 3m long x 0.4m diameter with an endurance of a 3mins at 2-3G but has an endurance of 1hour and accel of 10G.
Or the small Modular reactor (SMR) is 300dtons producing 50Mw against a nuclear sub which provides 300MW. That there is diminishing efficiency as the 300dton SMR gets smaller. Comparing it to a Radio-isotope Thermoelectric Generator (RTG used in many probes and satelites) of 470W at about 1/30 a dton or 0.5 a m^3.
Getting the Saturn V Apolo, Shuttle , Falcon Heavy, TianHe Programs Performance as a baseline for the TL8+ performance. This is because I want to help people run a 2050 Campaign where they are the growing number of Professional International Astronaughts that will be taking on the building of the first Permanent Commercial Space stations and Infrastructure - like in the Novel of Daniel Suarez Delta-V. Running games where PCs have to roll to succeed in using millions worth of Capital and high stakes projects.
Play Testing with Nicco and the guys the material, comparing the System as Written and the problems that comes up. Designing material to make the GMs prep small and making every investment of time. I see something that can be done and take action. Reminds me of Seth Skarkowskys reviews - its like its a Great Material, it has these problems, but these problems doesnt stop the material from being GREAT! Traveller has Great material, it not perfect, but being imperfect doesnt stop it from being Great.
We acknowledge the imperfection and act on it. It doesnt diminish the work and love that went into the work.