There is copy. And there is copy.
Not all reverse engineering is created equal.
It is wide spectrum. From one end of the spectrum, you have 'see concept, take concept', which can be called 'copy' only in most lose of sense. Most often cannot even prove the continuity of ideas, as often there has already been existing lines of thought. Common example - StG-44 and AK-47. Some say it is copy, even though internal workings completely different.
Then you have something like the Japanese copying of Jumo 004B - legend has it, from low-quality photograph.
Then - spying. Copying of blueprints.
Then, has copy for study - example Tu-4 and B-29.
Then, has copy, but for reference - such as - MiG-23s in China.
Then, licenced production, without a full range of ToT - from (low level of ToT) J-11, to high level of ToT - Su-30MKI.
Then, transfer of production line, though that is sometimes only extreme spectrum of above. Because licence production involves often transfer of part of line, and production of modern weapons barring most simple has so many different parts and components (Su-27 - 16 major bureaus for supplies alone, spread across time zones 6 hours apart) it is often completely unrealistic if not impossible to transfer entire line.
______________________________________
At every step of the way, like with design and prototype and testing and create manufacture protocol for item, there are so many ways to fail.
To completely walk through the entire process - congratulations, you have already mastered the industrial process from design to mass production.
_______________________________________
One way to look at it - the AK-47 mentioned above. Produced in half the world. Millions of copies. Design - is not complicated. Yet, you have anything from superior copies, designs which except parts of innards share next to nothing with original... From Czechs and Americans... to absolutely atrocious copies which barely can hit barn. When firing from inside. And if it does not burn soldier firing it is consider great success.
________________________________________
Even most crude "Copying" is involved process.
-Taking item apart (not all items designed for easy disassembly, without skilled technician. You can destroy your copy easily. I won't even know where to start taking apart aero-engine)
-Put back together (like first step. But harder. It is easy to break things)
-Create model of item (simply because can take apart and put back together does not make this part easy. Measuring everything is a pain - not all components in simple, easy to measure shapes. in fact most components are not)
-Understand model - this part can take years.
-Set design tolerance and requirements of parts - easier said than done.
-Create industrial process for every such part. Manufacture to the specs and tolerance that would make whole item work - again easier said than done. Often some components may have some specific requirements - such as shear strength or tensile at particular angle - that is very hard to anticipate
Congratulations. Now you have a (hopefully) working copy. Now you can begin testing and optimisation of production and assembly.
And this is talking about machines that can be taken apart and measured. When you start to deal with computers, IC boards, electronics... software (millions of lines of code for modern systems!) Complexity becomes ridiculous.
That is why to say 'is copy' without detailed context is almost completely pointless, and even if it is classical example of copy, a 'successful' copy ipso facto implies industrial and engineering expertise on par with original design and manufacturer. Again I remind gentlemen of AK-47. Manuals and blueprints practically given away, and yet quality of 'copy' - extremely wide spectrum.