Here's more things I would do as a CEO:
For Amiga, I would make few changes:
Amiga 1000, 2000 and 500 would be released as they did.
I'll axe Amiga 600/300 and 500+ and instead go full-on on Amiga 1200/3000 (with two Paula chips) in 1990 and Amiga 1200+/4000T in 1992.
Like I said, I would axe the CDTV, CD32 and C64GS, although if CDTV must exist, then I make Commodore sell it as Amiga CDTV from the get-go based on Amiga 1200 (with the Akiko chip and some improved tech from Amiga 3000/4000, so all "CD32" games would still exist) as an advanced multimedia computer.
How to rub Jack Tramiel in the face? By making Atari ST programs and games run on Amiga perfectly via emulator and add-on card (via clean-room reverse engineering like VTech did with Apple //). (Business is war, Jack.)
Amiga PowerPC would exist under my vision and use something like Amiga OS4 and I'll strike a deal with Silicon Graphics Inc. to licence their 3D RealityEngine tech to the Amiga and Commodore would provide Amiga compatibility for the SGI workstations via an emulator.
Of course, I would license the Workbench to Sony for PlayStation 1 in a similar vein to PS2 Linux project.
For additional stuff outside Amiga:
I would extend the Commodore 64's lifespan up until December 2004 (except in Japan, where it would still be produced up until to this day) and Plus/4 in 1994 and make the SID technology open for anyone to build upon (SidStation would be an official Commodore product). And I would licence the TED chip as a microcontroller to compete with Arduino, etc. with enhancements.
C64 would be a success in Japan under my vision, because of built-in add-on ROM to enable kanji characters without sacrificing compatibility with Western C64s, (at least the NTSC ones, although later models would get a PAL/NTSC switch).
And provide a capable toy kids' computer (a la VTech Genius IQ 128) as a competitor to VTech's toy computers