In this post, the OP asks about the integral,
$$I = \int_0^1\int_0^1\int_0^1\int_0^1\frac{1}{(1+x) (1+y) (1+z)(1+w) (1+ x y z w)} \ dx \ dy \ dz \ dw$$
I. User DavidH gave a beautiful (albeit long) answer in terms of the Nielsen generalized polylogarithm,
$$S_{n,p}(z) = \frac{(-1)^{n+p-1}}{(n-1)!\,p!}\int_0^1\frac{(\ln t)^{n-1}\big(\ln(1-z\,t)\big)^p}{t}dt$$
namely,
$$I = \tfrac32 S_{2,2}(-1)+\tfrac{11}{8} S_{1,3}(1)-S_{1,3}(-1) + \tfrac32 S_{3,1}(-1) \approx 0.223076$$
with the last addend tweaked by yours truly. A session with Mathematica shows that these explicitly are,
$$S_{3,1}(-1) = -\tfrac78\zeta(4) \\ S_{1,3}(1) = \zeta(4) \\ S_{2,2}(-1) = 2S_{1,3}(-1)-\tfrac18\zeta(4)$$
and,
$$S_{1,3}(-1) = \tfrac18\ln^3(2)\,\rm{Li}_1\big(\tfrac12\big)+\tfrac12\ln^2(2)\,\rm{Li}_2\big(\tfrac12\big)+\ln(2)\,\rm{Li}_3\big(\tfrac12\big)+\rm{Li}_4\big(\tfrac12\big)-\zeta(4)$$
Since $S_{1,3}(-1)$ and $S_{2,2}(-1)$ have a linear relation, then the integral can be simplified as,
$$\color{blue}{I = 2S_{1,3}(-1)+\tfrac14\zeta(4)}$$
Note that $\rm{Li}_n\big(\tfrac12\big)$ for $n=1,2,3$ have closed-forms.
II. User nospoon gave an equal but alternative form as,
$$I=\tfrac52\ln(2)\zeta(3)-\tfrac{11}{576}\pi^4-\tfrac1{2}\ln^2(2)\zeta(2)+\tfrac1{16}\ln^4(2)+\tfrac32\rm{Li}_4\big(\tfrac12\big)-A+\tfrac12B\\ \approx 0.223076$$
where
$$A = \int_0^1\frac{\rm{Li}_3(x)}{1+x}dx$$ $$B= \int_0^1\frac{\ln(1-x^2)\,\rm{Li}_2\big(\tfrac{1-x}2\big)}{x}dx$$
III. Question
After guessing on various candidate variables, is it true that the closed-forms of $A$ and $B$ are,
$$A = -4S_{2,2}(-1)+6S_{1,3}(-1) +\ln(2)\zeta(3) = 0.339545\dots$$ $$B = -\tfrac12S_{2,2}(-1)-2S_{1,3}(-1)-\tfrac38\ln(2)\zeta(3)+\tfrac14\ln^2(2)\zeta(2) = -0.1112606\dots$$
First solution: Modulo constants, the Nielsen polylog $S_{1,3}(-1)$ is equivalent to a special case of the celebrated Nielsen-Ramanujan integral:
Thus using the blue identity $\color{blue}{I = 2S_{1,3}(-1)+\tfrac14\zeta(4)}$ gives
Also one have $S_{2,2}(-1)$ equivalent to
Which leads to the
Second solution: All integrals proposed by OP are certain 4-admissible integrals (for reference, see here) which can be calculated via Multiple Zeta Values. The result is:
$A=\int_0^1 \frac{\text{Li}_3(x)}{x+1} \, dx=-2 \text{Li}_4\left(\frac{1}{2}\right)-\frac{3}{4} \zeta (3) \log (2)+\frac{\pi ^4}{60}-\frac{1}{12} \log ^4(2)+\frac{1}{12} \pi ^2 \log ^2(2)$
$B=\int_0^1 \frac{\text{Li}_2\left(\frac{1-x}{2}\right) \log \left(1-x^2\right)}{x} \, dx=-3 \text{Li}_4\left(\frac{1}{2}\right)-3 \zeta (3) \log (2)+\frac{47 \pi ^4}{1440}-\frac{1}{8} \log ^4(2)+\frac{1}{6} \pi ^2 \log ^2(2)$
So based on relations given by @nospoon it is proved again. Also here is an elementary solution of numerous integrals including $A,B$. Moreover this answers the OP's question i.e. verifying the correctness of $2$ formulas connecting $A, B, S_{p,q}$. In fact for these types of integrals a more powerful method exists, that is the
Third solution: Evidently, modulo a trivial $\log^n(2)$ term, the generalized integral $$I(n)=\int_{(0,1)^n} \frac{ \prod_1^n dx_i}{(1+\prod_1^n x_i)\prod_1^n (1+x_i)}$$ is equivalent to $$J(n)=\int_{(0,1)^n} \frac{\prod_1^n x_i \prod_1^n dx_i}{(1+\prod_1^n x_i)\prod_1^n (1+x_i)}$$ then tailed Euler sums $$\sum_{k=1}^\infty (\log(2)-\widetilde{H_k})^n (-1)^{(n+1)k}$$ then ordinary Euler sums (by Abel summation calculating partial sums of $(-1)^{(n+1)k}$ and taking difference on $(\log(2)-\widetilde{H_k})^n$, for at most $2$ times), then alternating (level $2$) MZVs via stuffle relations. Plugging in known special values of MZVs completes the evaluation of $I(4)$.
One obtain high-weight results similarly, say
$ I(6)=-33\zeta(\bar5,1)+60 \text{Li}_6\left(\frac{1}{2}\right)+30 \text{Li}_4\left(\frac{1}{2}\right) \log ^2(2)+60 \text{Li}_5\left(\frac{1}{2}\right) \log (2)+\frac{771 \zeta (3)^2}{64}+\frac{35}{4} \zeta (3) \log ^3(2)-\frac{29 \pi ^6}{360}+\frac{5 \log ^6(2)}{6}-\frac{5}{8} \pi ^2 \log ^4(2)$
$ I(7)=1729\zeta(\bar5,1)+\frac{35}{3} \pi ^2 \text{Li}_4\left(\frac{1}{2}\right)-3360 \text{Li}_6\left(\frac{1}{2}\right)-420 \text{Li}_4\left(\frac{1}{2}\right) \log ^2(2)-1680 \text{Li}_5\left(\frac{1}{2}\right) \log (2)-\frac{5397 \zeta (3)^2}{8}-\frac{315}{4} \zeta (3) \log ^3(2)+7 \pi ^2 \zeta (3) \log (2)-\frac{50813}{32} \zeta (5) \log (2)+\frac{1589281 \pi ^6}{362880}-\frac{14}{3} \log ^6(2)+\frac{175}{36} \pi ^2 \log ^4(2)+\frac{4739 \pi ^4 \log ^2(2)}{1440}$
Note that both $I(2k)$ and $I(2k+1)$ are generated by weight $2k$ constants, rather than of different weights (as expected). One may carry out Abel partial summation for both cases to see the reason. I cannot resist to give the following
Bonus: We have an associated magnificent series (have a try):